View
9
Download
0
Category
Preview:
Citation preview
The human cytidine deaminase APOBEC3C restricts retroviruses independent of editing
– A biochemical and structural analysis
vom Fachbereich Biologie der Technischen Universität Darmstadt
zur
Erlangung des akademischen Grades
eines Doctor rerum naturalium
genehmigte Dissertation von
Dipl. Biol. Henning Hofmann
aus Erbach/Odenwald
1. Referent: Prof. Dr. Gerhard Thiel
2. Referent: Prof. Dr. Carsten Münk
3. Referent: PD Dr. Arnulf Kletzin
Tag der Einreichung: 18. August 2010
Tag der mündlichen Prüfung: 3. Dezember 2010
Darmstadt 2011
D 17
Eidesstattliche Erklärung
Ich erkläre hiermit an Eides statt, dass ich die vorliegende Dissertation selbstständig
und nur mit den angegebenen Hilfsmitteln angefertigt habe. Ich habe bisher noch keinen
Promotionsversuch unternommen.
________________________
Henning Hofmann
Darmstadt, den 18. August 2010
Publications
Im Zuge der Promotion konnten folgende Arbeiten veröffentlicht werden: Vif of Feline Immunodeficiency Virus from domestic cats protects against APOBEC3 restriction factors from many felids J. Zielonka, D. Marino, H. Hofmann, N. Yuhki, M. Löchelt, and C. Münk J. Virol., 2010. 84: 7312-7324 Model structure of APOBEC3C reveals a binding pocket modulating ribonucleic acid interaction required for encapsidation B. Stauch*, H. Hofmann*, M. Perkovic, M. Weisel, F. Kopietz, K. Cichutek, C. Münk, and G. Schneider; Proc. Natl. Acad. Sci., Jul 21;106 (29):12079-84. 2009. *B.S. and H.H. contributed equally to this work High level expression of the anti-retroviral protein APOBEC3G is induced by influenza A virus but does not confer antiviral activity E-K. Pauli, M. Schmolke, H. Hofmann, C. Erhardt, E. Flory, C. Münk and S. Ludwig; Retrovirology, 2009, 6:38 Species-specific inhibition of APOBEC3C by the prototype foamy virus protein bet M. Perkovic, S. Schmidt, D. Marino, R. A. Russell, B. Stauch, H. Hofmann, F. Kopietz, B.-P. Kloke, J. Zielonka, H. Ströver, J. Hermle, D. Lindemann, V. K. Pathak, G. Schneider, M. Löchelt, K. Cichutek, and C. Münk; JBC, 2009. 284(9):5819-26 Teile dieser Arbeiten wurden in Chapter I der vorliegenden Arbeit verwendet. Abgesehen von folgenden Abbildungen in Chapter I wurden alle Daten von mir generiert: Fig. 1, Fig. 2, Fig. 3 (a+b), Fig. 7, Fig. 9 und Fig. S1 bis S5
Contents
Contents
Zusammenfassung 1
Summary 2 Introduction 3
Chapter I 15 Chapter II 30
Chapter III 41 Appendix 51
Zusammenfassung
1
Zusammenfassung
Die Fähigkeit des Humanen Immundefizienz Virus (HIV) in Zellen verschiedener Spezies zu
replizieren ist häufig abhängig von der An- oder Abwesenheit interagierender oder
einschränkender zellulärer Faktoren. Restriktionsfaktoren stellen einen Teil des
Abwehrmechanismus des Wirts gegen Pathogene dar. Die Familie der humanen APOBEC3 (A3)
Zytidin-Deaminasen ist Teil der intrinsischen Immunität und schützt höhere Säugetiere vor viralen
Pathogenen. A3 Proteine tragen eine oder zwei Kopien eines Zytidin-Deaminase Motivs und
können die retrovirale Replikation inhibieren, indem sie deren Genom während der reversen
Transkription deaminieren. Diese Inhibition wird von HIV durch das Vif-Protein überwunden,
welches den Abbau von A3s induziert und damit deren Verpackung in Virionen verhindert.
Obwohl ein Mitglied dieser humanen Proteinfamilie, APOBEC3C (A3C), hoch in vielen
menschlichem Geweben und in CD4+ T-Zellen exprimiert wird, kann A3C nur vif-defizientes SIV
(SIVΔvif), nicht aber HIVΔvif inhibieren. Ziel der vorliegenden Arbeit war es, 1.) die
unterschiedlichen Resistenzmechanismen von HIV und SIV gegenüber A3C zu definieren, 2.) den
Restriktionsmechanismus von A3C gegenüber SIV zu charakterisieren, und 3.) basierend auf
einem Strukturmodel Aminosäurereste zu identifizieren, die relevant für die antivirale Aktivität von
A3C sind. Mit der Hilfe eines dreidimensionalen Proteinmodels wurden Aminosäurereste
prognostiziert und experimentell identifiziert, die in die Dimerisierung des Proteins oder dessen
Verpackung in virale Partikel involviert sind. Die Ergebnisse zeigen, dass die Dimerisierung von
A3C notwendig für dessen antivirale Aktivität gegen SIV ist. Neben der Dimerisierung wird die
antivirale Aktivität durch Verpackung in virale Partikel reguliert. Die Verpackung von A3C beruht
auf einer Substratbindetasche, die distal des Zink-koordinierenden enzymatisch aktiven Zentrums
liegt. Diese Bindetasche vermittelt die RNA-abhängige Verpackung des Proteins in das knospende
Virus. Es konnte gezeigt werden, dass der Mechanismus der A3C-vermittelten Restriktion von
SIVΔvif nicht die Deaminierung virale ssDNA ist. Die experimentelle Untersuchung alternativer
Mechanismen zeigte, dass A3C weder virale RNA deaminiert, noch einen signifikanten
inhibitorischen Effekt auf die reverse Transkription von SIV hat. Daher ist anzunehmen, dass A3C
einen zusätzlichen, nicht bekannten Restriktionsmechanismus gegen SIV besitzt.
Ergebnisse dieser Arbeit zeigen weiterhin, dass die Resistenz von HIV gegen A3C überwunden
werden kann. Eine A3C vermittelte Restriktion von HIV wurde durch eine N- oder C-terminale
Fusion des viralen Proteins R (Vpr) von HIV-1 an A3C erreicht. Interessanter Weise wurden beide
Proteine, A3C und Vpr.3C, ins Virus verpackt und waren im gleichen viralen Kompartiment, dem
viralen Core, zu finden. Diese Daten weisen darauf hin, dass ein von HIV zusätzlich zu vif kodierter
Faktor der A3C Aktivität entgegenwirkt. Auf der Suche nach einem derartigen Faktor, konnte die
virale Integrase als ein Kandidat identifiziert werden.
Weitere Experimente sind notwendig, um die Interaktion von A3C mit HIV besser zu verstehen.
Jedoch implizieren die Ergebnisse dieser Arbeiten, dass eine Steigerung der antiviralen Aktivität
des ubiquitär exprimierten A3C die Replikation von HIV in Patienten eindämmen könnte und so
einen neuen Behandlungsansatz ermöglicht.
Summary
2
Summary
The capability of the Human immunodeficiency virus (HIV) to replicate in cells of different species
depends on the presence or absence of interacting or restricting cellular factors. The restriction
factors are part of the host’s defense mechanism against pathogens. The family of human
APOBEC3 (A3) cytidine deaminases forms part of the intrinsic immunity and protects placental
mammals from viral pathogens. A3 proteins have one or two copies of a cytidine deaminase motif
and can restrict retroviral replication by deamination of their genomes during reverse transcription.
HIV counteracts this inhibition by the Vif-protein, which induces the destruction of A3s and thereby
prevents its incorporation into virions.
Although one member of this human protein family, APOBEC3C (A3C), is highly expressed in
many human tissues and in CD4+ T cells, A3C does not restrict vif-deficient HIV (HIV Δvif) but is
active against SIV Δvif. This study aimed to characterize i) the different mechanisms of resistance
of HIV and SIV against A3C, ii) the restriction mechanism of A3C against SIV, and iii) residues
crucial for the antiviral activity of A3C based on a structural model. With the help of a three-
dimensional protein model of A3C, amino acids residues were predicted and finally experimentally
identified that are involved in protein dimerization and that mediate the incorporation of A3C into
virions. The result showed that dimerization of A3C is essential for antiviral activity against SIV.
Beside dimerization, the antiviral activity is regulated by encapsidation into viral particles.
Encapsidation of A3C depends on the substrate binding-pocket distal from the zinc-coordinating
enzymatic centre. This binding pocket mediates the RNA-dependent incorporation of the protein
into budding viral particles. It was found that A3C restricts SIV Δvif without cytidine deamination of
viral ssDNA. Therefore, alternative mechanisms were investigated. In this study the experiments
showed clearly that A3C does neither deaminates viral RNA nor has a significant inhibitory effect
on the reverse transcription of SIV. Thus, it is likely that A3C has a so far unknown restriction
mechanism against SIV Δvif.
Results of this study also show that the resistance of HIV against A3C can be overcome. An A3C
mediated inhibition of HIV was achieved by fusing the Viral Protein R (Vpr) of HIV-1 either to the C-
or N- terminus of A3C. Interestingly both, A3C and Vpr.3C were incorporated into viral particles and
found to be localized in the same viral compartment, the viral core. These data indicate that HIV
encodes a factor additional to vif to counteract the A3C activity. In search for such a viral factor, the
viral integrase was identified as a candidate inhibitor of A3C.
Additional experiments are required for a better understanding of the interaction of A3C with HIV.
However, the results here implicate that enhancing the antiviral activity of the ubiquitously
expressed A3C would likely repress HIV-1 replication in patients and thus should be considered as
a novel approach for treatment.
Introduction
3
Introduction
In the year 1981 a new epidemic in human history, the acquired immunodeficiency syndrome
(AIDS), appeared and 3 years later it was found that a novel retrovirus was the aetiologic agent.
Infection with the Human immunodeficiency virus type 1 (HIV-1) causes a loss of CD4 positive T-
cells, thereby weakening the human immune system and enhancing the risk of future opportunistic
infections. HIV belongs to a large family of retroviruses including simian immunodeficiency viruses
(SIVs). HIV-1 developed at least three times in humans by three independent transmissions from
an SIV infected chimpanzee to humans forming the HIV-1 clades M, N and O. The transmission of
HIV between humans mostly occurs through unsafe sex and from contaminated needles from
injection drug users. In 2008 an estimated number of 33.4 million people (UN-AIDS; WHO) in the
world were infected with HIV while another 25 million people already died of this disease. Most of
the infected people, in numbers 22.4 million, live in Sub-Saharan Africa. Although the expanding
understanding of the HIV life cycle has led to the development of antiretroviral drugs like inhibitors
of the viral enzymes protease and reverse transcriptase, there is no vaccine or cure for HIV and
AIDS available. Currently HIV infected people are treated with the highly active antiretroviral
therapy (HAART), a therapy that can prolong the life of infected people to a mostly normal lifespan.
Since the discovery of HIV, much work has been done to understand the replication of HIV in
humans and to investigate the interactions of the virus with its host. Understanding the interaction
of the virus with host cells restriction factors is a promising new aspect in the HIV research. Host
cell restriction factors of non-human species are able to prevent viral infections by blocking different
steps of the viral life cycle. Adapted viruses such as HIV to human overcome these restriction
factors through accessory gene products that counteract the antiviral restriction. The
characterization of the molecular basis of the interaction of viral with cellular determinants can
highlight new targets for HIV therapy. This study focuses on the molecular understanding of the
interaction between the cellular restriction factor APOBEC3C (A3C) and HIV or SIV.
Introduction
4
A| RETROVIRUSES
Retroviruses are enveloped RNA viruses forming the family of Retroviridae. They are diploid as
each virion contains two copies of its ssRNA genome. During replication the retroviral genome is
reverse transcribed into DNA by the viral enzyme reverse transcriptase (RT). The viral integrase
(IN) catalyzes the integration of viral DNA into the host cell genome. The integration of the proviral
DNA results in a lifelong persistent infection of the host.
The basic genome of retroviruses is composed of three genes: gag (group specific antigen), pol
(polymerase) and env (envelope) (Fig. A-1). The provirus of all replication competent retroviruses is
flanked by LTRs (long terminal repeats).
Fig. A-1: Schematic overview of the genome organization of a simple retrovirus. LTRs are depicted in light
grey, in dark grey the genes gag, pol and env are shown with their protein products (white letters). The primer
binding site (PBS) is shown as small bar. (Figure modified from Malim, M., 2009)
The gag gene encodes for structural proteins of the virus: matrix (MA), capsid (CA) and
nucleocapsid (NC). The genetic information for the viral enzymes protease (PR), reverse
transcriptase (RT), RNaseH (RH) and integrase (IN) is located on the pol gene. The two envelope
proteins encoded by the env gene are the surface glycoprotein (SU) and the transmembrane
glycoprotein (TM).
Fig. A-2: Composition of a mature retroviral particle. The dimeric RNA genome together with a cellular tRNA
and the viral enzymes IN and RT is surrounded by CA proteins build the viral core. A layer of MA proteins
separates the core from the envelope proteins (SU and TM).
Introduction
5
The viral particle contains a dimer of the viral ssRNA as well as host cell RNA (Fig.2). This host cell
RNA is a transfer RNA (tRNA) which is bound to the PBS of the viral RNA. The tRNA is required for
the initiation of reverse transcription. Some viral proteins are associated with the RNA, such as NC
that coats the RNA and the viral enzymes IN and RT. A layer of capsid (CA) proteins forms the viral
core casing this ribonucleic acid-protein (RNP)-complex. The capsid proteins are surrounded by a
layer of MA-proteins building a border between CA and envelope proteins. The envelope of
retroviruses is build of two, non-covalently linked proteins, a TM protein and the heavily
glycosylated SU protein which is the ligand for the cellular receptors and responsible for the
recognition of the target cell. The retroviral life cycle will be illustrated using the Human
immunodeficiency virus (HIV) by way of example.
B| HUMAN IMMUNODEFICIENCY VIRUS (HIV)
HIV is a complex retrovirus that belongs to the genus of Lentiviruses. In addition to the basic
retroviral genes gag, pol and env the HIV genome encodes for accessory genes as shown in Fig.
B-1. The gene products of these accessory genes play important roles in the HIV life cycle as they
either directly support viral replication like (Tat or Rev) (1-3) or counteract host cell restriction
factors (like Vif or Vpu) (4, 5). The biological relevant functions of Vpr and Nef are less clear.
Fig. B-1: Schematic representation of the HIV-1 genome flanked by the 5´ and 3´ LTR regions (light grey).
Gag, pol and env genes are shown in dark grey. HIV encodes the accessory genes vif (light pink), vpr (green),
tat (dark pink), rev (purple), vpu (blue) and nef (brown) in addition to the basic retroviral genes. The PBS is
shown as small bar. (Figure modified from (6)).
Cells of the human immune system expressing the cellular receptor CD4 (CD4+ cells), like T-helper
cells, macrophages or dentritic cells, are the target cells for HIV infection. HIV infection can be
separated into three steps: the acute infection, a latency phase and the progression of the Acquired
Immunodeficiency Syndrome (AIDS) (Fig B-2). In early HIV infection the number of CD4+ cells
dramatically drops down, either as a result of apoptosis triggered by the viral proteins Nef and Vpr
or as a result of CD8+ T-cell recognition and killing (reviewed in (7) and (8)). During this phase of
acute viral infection the virus spreads, being detectable through a high viral load in the plasma of
patients. The phase of acute infection lasts for several weeks before the next stage of HIV
infection, the latency phase, begins which in average last for years. In this phase the viral load
decreases, due to a strong defence of the host immune system. Viral latency lasts until the number
Introduction
6
of CD4+ T-cells drops below a limit where cell-mediated immunity gets lost and the progression of
AIDS begins (reviewed in (7)). Loss of CD4+ T-cell mediated immunity enhances the risk of future
opportunistic infections.
Fig. B-2: The course of HIV infection. The acute infection is characterized by a high plasma viremia (red line,
top) and a decrease in CD4+ cells (green line, bottom). After the latency phase displayed through a recovery
of the CD4+ cells and a lowered viral load, the amount of CD4+ drops down and the number of viruses in the
plasma increases again resulting in AIDS progression. (Figure modified from (8)).
To establish preventive or therapeutic measures against HIV infections a closer look on the viral
replication is necessary (Fig. B-3). The HIV life cycle starts with the infection of CD4+ target cells,
either as a free virus particle or transmitted by dendritic cells within the virologic synapse. The
binding of the HIV surface glycoprotein gp120 to a cellular CD4 receptor induces a rearrangement
of the envelope complex allowing another domain of gp120 to attach to a chemokine receptor,
either CCR5 or CXCR4 (9). This interaction allows gp41 to penetrate the cell membrane resulting
in the fusion with the viral membrane (reviewed in (10)). After fusion the viral core is released into
the cytoplasm. Prior to reverse transcription uncoating of the viral core occurs. Binding of the tRNA
to the PBS is required for reverse transcription and the reaction is catalyzed through the viral
reverse transcriptase (11). Viral ssRNA thereby gets transcribed into dsDNA. The nuclear import of
the cDNA is triggered by a pre-integration complex that is transported into the nucleus via nuclear
pores. In the nucleus the IN catalyzes the integration of the viral DNA genome into the
chromosomal genome of the host cell resulting in the provirus formation (12). Transcription of the
integrated provirus is initiated by cellular transcription factors that interact with the viral promoter in
the 5´ LTR region.
Introduction
7
Fig B-3: Replication cycle of HIV. Important steps are highlighted with white boxes. (Figure modified from (7))
The first viral transcripts are multiply spliced and exported from the nucleus to the cytoplasm where
the early gene expression of Tat, Rev and Nef occurs (13). Tat and Rev both have a nuclear
localization signal that targets them to the nucleus. Tat enhances the proviral transcription (2, 3)
and Rev binds to unspliced or single spliced proviral RNA and thereby promotes their transport to
the cytoplasm (1). In the cytoplasm the late gene products like Vif, Vpu, Vpr and Env are translated
from the single spliced RNAs, whereas the unspliced RNA is translated into the gag-pol
polyprotein. In order to facilitate the assembly of newly produced viral particles, two molecules of
the unspliced viral RNA interact with the NC of gag-pol polyproteins which become anchored in the
plasma membrane (reviewed in (14)). Through the recruitment of the cellular ESCRT (= endosomal
sorting complex required for transport) proteins budding of the virion from the cellular membrane
takes place (reviewed in (15)). Generation of an infectious viral particle (maturation) occurs after
virus release through cleavage of the gag-pol polyproteins by the viral protease (14).
Certain steps in the viral life cycle are targets of known host cell restriction factors. An overview of
restriction factors with special focus on APOBEC3 proteins is given in the following.
C| RESTRICTION FACTORS AND APOBEC3 PROTEINS
Some cells are non-permissive for specific viruses. Non-permissivity is based on the absence of
essential co-factors or due to the presence of factors restricting viral replication in these cells.
These restriction factors can block viruses at different steps of the viral life cycle and adapted
viruses have evolved strategies to circumvent these restrictions. HIV-1 is able to counteract
Introduction
8
different restriction factors with its own accessory proteins. A lack of such a protein inhibits
replication in non-permissive cells, but is not inhibitory in permissive cells.
One example for a restriction factor counteracted by a viral protein is tetherin (also known as BST-
2 or CD317). This protein was first described by Neil et al 2008 (16). Tetherin-expressing cells
were non-permissive for HIV∆vpu, whereas vpu-deleted viruses could replicate in permissive cells
that did not express tetherin. In the absence of the viral protein U (Vpu) HIV particles are tethered
at the cellular membrane and viral release is impaired. Although it was observed, that Vpu down-
regulates cellular tetherin levels (17-19) and that both proteins show physical interaction (17), it
remains unclear by which exact mechanism Vpu counteracts the restriction of tetherin. Additionally,
lentiviruses that do not encode Vpu can also counteract the tetherin mediated restriction. In the
case of HIV-2, a Vpu-like activity is encoded in its Env glycoprotein (20-22) and the Simian
immunodeficiency virus from African Green Monkeys (SIVagm) and from Macaques (SIVmac) are
able to circumvent the tetherin restriction via their Nef proteins (23, 24). Another example for a
complex and diverse virus-host cell interaction is the family of APOBEC3 restriction proteins, which
are subject of this study.
Fig. C-1: Schematic representation of the human APOBEC protein family. The Zn2+ -coordinating motifs are
shown in red and the crucial, Zn2+ - binding amino acids are highlighted. (Figure taken from (25))
The family of APOBEC3 (apolipoprotein B mRNA-editing enzyme, catalytic polypeptide-like 3)
restriction factors was identified through the observation, that ∆vif HIV was able to infect the cell
line CEM-S (permissive) but not the CEM-SS cell line (non-permissive) (26). This observation led
to the idea of a restriction factor that is expressed in the non-permissive CEM-SS cells and can be
inactivated by the Vif protein of HIV, because wildtype (wt) HIV was able to replicate in CEM-SS
cells. Screening of the expression profiles of these near-isogenic cell lines resulted in the
identification of the protein APOBEC3G (A3G). This cytidine deaminase was expressed in CEM-SS
cells, but not in the permissive CEM-S cells. A3G is a member of the human APOBEC cytidine
deaminase family that consists of 11 members shown in Figure C-1.
Introduction
9
Although this family is very homologous sharing one or two Zn2+ coordinating domains the proteins
exhibit different enzymatic activities. APOBEC1 (A1), the founder of this protein family, is described
as a RNA deaminase expressed in gastrointestinal tissue (27). A1 deaminates the cytosine 6666 in
apolipoprotein mRNA resulting in an in-frame stop codon (28, 29). The truncated and the full-length
protein have different functions in the lipid metabolism. The DNA editing activity of the activation
induced deaminase (AID) in B-cells leads to somatic hypermutation and class-switch recombination
of antibodies, and thus plays a key role in the maturation of the natural antibody response (30-32).
Whereas APOBEC2 was shown to be necessary for muscle development (33), the function of
APOBEC4 (A4) is still unknown. The human APOBEC3 (A3) family has been described as antiviral
proteins restricting retroviruses and retrotransposons. The antiviral activity of A3 proteins is
ascribed to the deamination of cytidine to uridine of single-stranded viral DNA formed during
Fig. C-2: APOBEC-dependent restriction of HIV infection. (a) In cells infected with wt HIV A3 proteins are
excluded from budding viral particles due to Vif-induced proteasomal degradation. (b) A3 proteins present in
producer cells get incorporated into Vif-deficient viral particles and deaminate the viral genome after infection
of a target cell. (Figure taken from (25))
reverse transcription (34-37). This DNA editing either leads to non-functional viral genes as a result
of extensive G to A hypermutation on the coding strand of viral DNA or to degradation of viral DNA
Introduction
10
by the DNA base repair enzymes uracil DNA glycosylase and apurinic-apyrimidinc endonuclease
(38). A prerequisite for this antiviral activity is the incorporation of the A3 proteins from the producer
cell into budding virions. In most cases, A3s being present in viral particles deaminate the viral
genome after infection of a target cell (Fig. C-2b). Besides editing other mechanisms of A3
mediated viral restriction, like inhibition of integration or reverse transcription have been discussed
(39-43). The Vif protein defeats the activity of A3 proteins by binding them and recruiting an E3
ubiquitin ligase complex (44). A3 proteins get poly-ubiquitylated and subsequent degraded by the
proteasome (5, 45)(Fig C-2a). Thus A3 proteins are no longer incorporated into viral particles and
the virus remains fully infectious.
Interestingly, the Vif – APOBEC3 interaction is often species-specific. Human APOBEG3G (hA3G)
for example is recognized by the VifHIV-1, but the Vif protein of the Simian immunodeficiency virus
Fig. C-3: A model for the evolutionary origin of human APOBEC3 genes located on chromosome 22. The
ancestor domains of A3A, A3C and A3H are colored in green, orange and blue, respectively. Black lines
indicate duplications within the original segment, red lines represent duplication events resulting in a
duplicated segment and grey lines indicate no change. Black crosses symbolize gene deletions. (Figure taken
from (46))
from African Green Monkeys (SIVagm) is not capable to induce the degradation of hA3G (47). Vice
versa, agmA3G is only recognized by VifSIVagm, but not by VifHIV-1. The restriction phenotype of the
individual human A3 proteins highly differs from each other. The most prominent member of the
human APOBEC3 protein family is A3G that exhibits antiviral activity against ∆vif HIV, many other
retroviruses and retroelements (for review (25); see table 1 and references within). A3G has two
Introduction
11
Zn2+ coordinating motifs, a C-terminal domain (CTD) and a N-terminal domain (NTD), with different
activities (48-50). While the NTD is required for its incorporation into viral particles, the CTD
catalyzes the deamination of the viral genome. Another member of the A3 protein family, A3C,
contains only one Zn2+ coordinating domain and is not active against ∆vif HIV, but shows a robust
restriction of ∆vif SIV (51) and Long Interspersed Element-1 (LINE-1) retrotransposons (52).
Whether A3 proteins have one or two domains is a result of gene-duplications and –deletions from
3 ancestor A3 proteins A3Z1, A3Z2 and A3Z3 (46)(Fig. C-3).
Interestingly, the ancestor domain of A3C, A3Z2, is present in each of the two-domain APOBEC3
proteins, showing that A3C is an evolutionary highly conserved protein. Additionally, it is broadly
expressed in many human tissues (53), a variety of cancer cell lines (54) and is up regulated in
CD4+ cells, the target cells of HIV (55). As HIV is able circumvent the A3C mediated restriction also
in absence of its Vif protein (51), there is likely an additional factor encoded by HIV that can
counteract the antiviral activity of A3C. Through these observations A3C has become an interesting
candidate to explore virus-host interactions.
D| SCOPE OF THIS WORK
This study aimed to understand the interaction of the restriction factor APOBEC3C with HIV and
SIV. Therefore residues of A3C crucial for its antiviral activity were identified based on a structural
model, the different restriction phenotypes of A3C against HIV and SIV and putative mechanisms
of antiviral restriction were analyzed. In the different chapters the following questions were
answered or aimed to answer:
CHAPTER I: Which amino acids and protein structures are crucial for the antiviral activity of
A3C against SIV ∆vif?
CHAPTER II: Why is A3C antiviral against SIV ∆vif, but not against HIV ∆vif?
CHAPTER III: What is the restriction mechanism of A3C against SIV ∆vif and is the Zn2+-
coordinating domain the enzymatic active site of the protein?
References
1. Pollard VW & Malim MH (1998) The HIV-1 Rev protein. Annu. Rev. Microbiol. 52, 491-532.
2. Dayton AI et al. (1986) The trans-activator gene of the human T cell lymphotropic virus type III is required for replication. Cell 44, 941-947.
3. Fisher AG et al. (1986) The trans-activator gene of HTLV-III is essential for virus replication. Nature 320, 367-371.
4. Neil SJ, Zang T & Bieniasz PD (2008) Tetherin inhibits retrovirus release and is antagonized by HIV-1 Vpu. Nature 451, 425-430.
5. Yu X et al. (2003) Induction of APOBEC3G ubiquitination and degradation by an HIV-1 Vif-Cul5-SCF complex. Science 302, 1056-1060.
6. Malim MH (2009) APOBEC proteins and intrinsic resistance to HIV-1 infection. Philos. Trans. R. Soc. Lond B Biol. Sci. 364, 675-687.
7. Stevenson M (2003) HIV-1 pathogenesis. Nat. Med. 9, 853-860.
Introduction
12
8. Simon V, Ho DD & Abdool KQ (2006) HIV/AIDS epidemiology, pathogenesis, prevention, and treatment. Lancet 368, 489-504.
9. Clapham PR & McKnight A (2002) Cell surface receptors, virus entry and tropism of primate lentiviruses. J. Gen. Virol. 83, 1809-1829.
10. Sierra S, Kupfer B & Kaiser R (2005) Basics of the virology of HIV-1 and its replication. J. Clin. Virol. 34, 233-244.
11. Harrich D & Hooker B (2002) Mechanistic aspects of HIV-1 reverse transcription initiation. Rev. Med. Virol. 12, 31-45.
12. Van MB & Debyser Z (2005) HIV-1 integration: an interplay between HIV-1 integrase, cellular and viral proteins. AIDS Rev. 7, 26-43.
13. Purcell DF & Martin MA (1993) Alternative splicing of human immunodeficiency virus type 1 mRNA modulates viral protein expression, replication, and infectivity. J. Virol. 67, 6365-6378.
14. Freed EO (2001) HIV-1 replication. Somat. Cell Mol. Genet. 26, 13-33.
15. Bieniasz PD (2009) The cell biology of HIV-1 virion genesis. Cell Host. Microbe 5, 550-558.
16. Neil SJ, Zang T & Bieniasz PD (2008) Tetherin inhibits retrovirus release and is antagonized by HIV-1 Vpu. Nature 451, 425-430.
17. Douglas JL et al. (2009) Vpu directs the degradation of the human immunodeficiency virus restriction factor BST-2/Tetherin via a {beta}TrCP-dependent mechanism. J. Virol. 83, 7931-7947.
18. Miyagi E, Andrew AJ, Kao S & Strebel K (2009) Vpu enhances HIV-1 virus release in the absence of Bst-2 cell surface down-modulation and intracellular depletion. Proc. Natl. Acad. Sci. U. S. A 106, 2868-2873.
19. Van DN et al. (2008) The interferon-induced protein BST-2 restricts HIV-1 release and is downregulated from the cell surface by the viral Vpu protein. Cell Host. Microbe 3, 245-252.
20. Abada P, Noble B & Cannon PM (2005) Functional domains within the human immunodeficiency virus type 2 envelope protein required to enhance virus production. J. Virol. 79, 3627-3638.
21. Bour S & Strebel K (1996) The human immunodeficiency virus (HIV) type 2 envelope protein is a functional complement to HIV type 1 Vpu that enhances particle release of heterologous retroviruses. J. Virol. 70, 8285-8300.
22. Bour S, Schubert U, Peden K & Strebel K (1996) The envelope glycoprotein of human immunodeficiency virus type 2 enhances viral particle release: a Vpu-like factor? J. Virol. 70, 820-829.
23. Jia B et al. (2009) Species-specific activity of SIV Nef and HIV-1 Vpu in overcoming restriction by tetherin/BST2. PLoS. Pathog. 5, e1000429.
24. Zhang F et al. (2009) Nef proteins from simian immunodeficiency viruses are tetherin antagonists. Cell Host. Microbe 6, 54-67.
25. Holmes RK, Malim MH & Bishop KN (2007) APOBEC-mediated viral restriction: not simply editing? Trends Biochem. Sci. 32, 118-128.
26. Sheehy AM, Gaddis NC, Choi JD & Malim MH (2002) Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein. Nature 418, 646-650.
27. Nakamuta M et al. (1995) Alternative mRNA splicing and differential promoter utilization determine tissue-specific expression of the apolipoprotein B mRNA-editing protein (Apobec1) gene in mice. Structure and evolution of Apobec1 and related nucleoside/nucleotide deaminases. J. Biol. Chem. 270, 13042-13056.
28. Knott TJ et al. (1986) Complete protein sequence and identification of structural domains of human apolipoprotein B. Nature 323, 734-738.
Introduction
13
29. Yang CY et al. (1986) Sequence, structure, receptor-binding domains and internal repeats of human apolipoprotein B-100. Nature 323, 738-742.
30. Martin A et al. (2002) Activation-induced cytidine deaminase turns on somatic hypermutation in hybridomas. Nature 415, 802-806.
31. Muramatsu M et al. (2000) Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell 102, 553-563.
32. Okazaki IM et al. (2002) The AID enzyme induces class switch recombination in fibroblasts. Nature 416, 340-345.
33. Sato Y et al. (2010) Deficiency in APOBEC2 leads to a shift in muscle fiber type, diminished body mass, and myopathy. J. Biol. Chem. 285, 7111-7118.
34. Bishop KN et al. (2004) Cytidine deamination of retroviral DNA by diverse APOBEC proteins. Curr. Biol. 14, 1392-1396.
35. Lecossier D, Bouchonnet F, Clavel F & Hance AJ (2003) Hypermutation of HIV-1 DNA in the absence of the Vif protein. Science 300, 1112.
36. Mangeat B et al. (2003) Broad antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. Nature 424, 99-103.
37. Zhang H et al. (2003) The cytidine deaminase CEM15 induces hypermutation in newly synthesized HIV-1 DNA. Nature 424, 94-98.
38. Yang B et al. (2007) Virion-associated uracil DNA glycosylase-2 and apurinic/apyrimidinic endonuclease are involved in the degradation of APOBEC3G-edited nascent HIV-1 DNA. J. Biol. Chem. 282, 11667-11675.
39. Bishop KN, Holmes RK & Malim MH (2006) Antiviral potency of APOBEC proteins does not correlate with cytidine deamination. J. Virol. 80, 8450-8458.
40. Guo F et al. (2006) Inhibition of formula-primed reverse transcription by human APOBEC3G during human immunodeficiency virus type 1 replication. J. Virol. 80, 11710-11722.
41. Holmes RK, Koning FA, Bishop KN & Malim MH (2007) APOBEC3F can inhibit the accumulation of HIV-1 reverse transcription products in the absence of hypermutation. Comparisons with APOBEC3G. J. Biol. Chem. 282, 2587-2595.
42. Mbisa JL et al. (2007) Human immunodeficiency virus type 1 cDNAs produced in the presence of APOBEC3G exhibit defects in plus-strand DNA transfer and integration. J. Virol. 81, 7099-7110.
43. Mbisa JL, Bu W & Pathak VK (2010) APOBEC3F and APOBEC3G inhibit HIV-1 DNA integration by different mechanisms. J. Virol. 84, 5250-5259.
44. Yu Y et al. (2004) Selective assembly of HIV-1 Vif-Cul5-ElonginB-ElonginC E3 ubiquitin ligase complex through a novel SOCS box and upstream cysteines. Genes Dev. 18, 2867-2872.
45. Marin M, Rose KM, Kozak SL & Kabat D (2003) HIV-1 Vif protein binds the editing enzyme APOBEC3G and induces its degradation. Nat. Med. 9, 1398-1403.
46. LaRue RS et al. (2008) The artiodactyl APOBEC3 innate immune repertoire shows evidence for a multi-functional domain organization that existed in the ancestor of placental mammals. BMC. Mol. Biol. 9, 104.
47. Mariani R et al. (2003) Species-specific exclusion of APOBEC3G from HIV-1 virions by Vif. Cell 114, 21-31.
48. Hache G, Liddament MT & Harris RS (2005) The retroviral hypermutation specificity of APOBEC3F and APOBEC3G is governed by the C-terminal DNA cytosine deaminase domain. J. Biol. Chem. 280, 10920-10924.
Introduction
14
49. Navarro F et al. (2005) Complementary function of the two catalytic domains of APOBEC3G. Virology 333, 374-386.
50. Gooch BD & Cullen BR (2008) Functional domain organization of human APOBEC3G. Virology 379, 118-124.
51. Yu Q et al. (2004) APOBEC3B and APOBEC3C are potent inhibitors of simian immunodeficiency virus replication. J. Biol. Chem. 279, 53379-53386.
52. Muckenfuss H et al. (2006) APOBEC3 proteins inhibit human LINE-1 retrotransposition. J. Biol. Chem. 281, 22161-22172.
53. Koning FA et al. (2009) Defining APOBEC3 expression patterns in human tissues and hematopoietic cell subsets. J. Virol. 83, 9474-9485.
54. Chiu YL & Greene WC (2008) The APOBEC3 cytidine deaminases: an innate defensive network opposing exogenous retroviruses and endogenous retroelements. Annu. Rev. Immunol. 26, 317-353.
55. Refsland EW et al. (2010) Quantitative profiling of the full APOBEC3 mRNA repertoire in lymphocytes and tissues: implications for HIV-1 restriction. Nucleic Acids Res. 38, 4274-4284.
Chapter I
15
CHAPTER I
Model Structure of APOBEC3C Reveals a Binding Pocket
Modulating RNA Interaction Required for Encapsidation
Benjamin Stauch1,2*, Henning Hofmann3,4*, Mario Perković3,4, Martin Weisel1, Ferdinand
Kopietz3, Klaus Cichutek3, Carsten Münk3,4†, Gisbert Schneider1†
1Johann Wolfgang Goethe-University Frankfurt, Chair for Chem- and Bioinformatics, Siesmayerstr. 70, 60323
Frankfurt am Main, Germany. 2present address: EMBL Heidelberg, Structural and Computational Biology Unit,
Meyerhofstr. 1, 69117 Heidelberg, Germany. 3Paul-Ehrlich-Institute, Division of Medical Biotechnology, Paul-
Ehrlich-Str. 51-59 63225, Langen, Germany. 4present address: Heinrich-Heine-University Düsseldorf, Clinic
for Gastroenterology, Hepatology and Infectiology, Building 23.12_U1 Room 81, Moorenstr. 5, 40225
Düsseldorf, Germany.
* BS And HH contributed equally to this work † Corresponding authors: CM and GS
ABSTRACT
Human APOBEC3 (A3) proteins form part of
the intrinsic immunity to retroviruses.
Carrying one or two copies of a cytidine
deaminase motif, A3s act by deamination of
retroviral genomes during reverse
transcription. HIV-1 overcomes this inhibition
by the Vif-protein, which prevents
incorporation of A3 into virions. In this study,
we modeled and probed the structure of
APOBEC3C (A3C), a single-domain A3 with
strong anti-lentiviral activity. The three-
dimensional protein model was used to
predict the effect of mutations on antiviral
activity, which was tested in a Δvif SIV-
reporter-virus assay. We found that A3C-
activity requires protein dimerization for
antiviral activity against SIV. Furthermore, by
using a structure-based algorithm for
automated pocket extraction we detected a
putative substrate binding-pocket of A3C
distal from the zinc-coordinating deaminase
motif. Mutations in this region diminished
antiviral activity by excluding A3C from
virions. We found evidence that the small
5.8S RNA specifically binds to this locus and
mediates incorporation of A3C into virus
particles.
KEY WORDS Bioinformatics, immunodeficiency, protein-
protein interaction, protein structure,
retrovirus
Chapter I
16
INTRODUCTION
One of the best-characterized cellular
proteins efficiently restricting Human
immunodeficiency virus type-1 (HIV-1) is
APOBEC3G (A3G) (1). Encapsidation of A3G
in HIV-1 virus particles leads to deamination
of cytosine residues to uracil in growing
single-stranded DNA during reverse
transcription (2-6). A3G has additional, still ill-
defined antiviral activities (7). HIV-1 uses the
viral infectivity factor (Vif) to prevent or
reduce incorporation of A3G into progeny
virions (4, 8, 9).
The human genome contains seven
APOBEC3 (A3) genes which can be
classified according to the presence of the
Z1, Z2 and Z3 zinc-coordinating motifs (10,
11). Z2, the A3C family, consists of A3C, the
C- and N-terminal domains of A3DE and
A3F, and the N-terminal domains of A3B and
A3G. The Z1 group, the A3A family, contains
A3A and the C-terminal domains of A3B and
A3G. A3H represents the Z3 zinc-finger
domain. Accordingly, A3B, A3G, A3DE and
A3F have two domains, while A3A, A3C and
A3H possess only one domain (12). In the
human A3 locus there is evidence for gene
expansion, and it was speculated that
duplications of single-domain genes led to
the evolution of the two-domain A3s (13).
Phylogentic analysis of primate and non-
primate antiviral cytidine deaminases showed
that in the early evolution of mammals genes
for A3C (Z2), A3A (Z1) and A3H (Z3) were
already present (11). Among these antetype
A3s, human A3A and most variants of A3H
are not antiviral against HIV (14-16). While
A3C is packaged into Δvif HIV with a weak
antiviral effect (17), A3C is a strong inhibitor
of Δvif SIV (18). The study of A3C gains
further importance from the fact that for A3s
until now only structures of Z1-derived
domains, e.g. A3G-CD, have been solved
experimentally. Notably, both A3C and the
still ill-defined N-terminal domain of A3G are
of type Z2. A study by Bourara et al. (19)
shows that in target cells A3C can induce
limited G-to-A mutations in HIV. These
mutations do not block viral replication, but
rather contribute to viral diversity.
Fundamental biochemical aspects of the A3
protein structure and their relevance for
antiviral activity are still a matter of
discussion. Here, we performed comparative
protein modeling of A3C, and assessed the
model using A3C-mutants in the SIVagm
system. This study provides a first structural
basis for rational antiviral intervention
targeting A3C. We found evidence that A3C
dimerization is critical for antiviral activity.
Furthermore, we found a previously
undescribed cavity in A3C that is similar to
nucleic acid binding pockets of known
enzymes. A point mutation near the pocket
diminishes encapsidation of A3C and
reduces 5.8S-RNA-binding. We hypothesize
that the natural substrate of this pocket of
A3C is a nucleic acid, possibly mediating its
incorporation into the virion by interaction
with nucleocapsid protein.
Chapter I
17
RESULTS
Comparative modeling of A3C
Structures of APOBEC2 (A2) and A3G, C-
terminal domain (A3G-CD), have been
solved experimentally (20-23). A2 crystallizes
as a homo-tetramer (PDB identifier: 2NYT,
2.5 Å resolution), composed of two outer and
two inner monomers, forming a dimer of
dimers, each of whose β-strands form an
extended β-sheet. Each monomer possesses
one copy of the conserved deaminase motif
H-X-E-X23-28-C-X2-4-C coordinating one
catalytic Zn2+-ion. While the overall
conformations of the inner and outer
monomers − chains A and C, B and D − differ
only slightly (0.2-1.1 Å pairwise root mean
square deviation (RMSD)), the orientation of
E60 with respect to the Zn2+ differs
remarkably, possibly representing a
molecular switch between the active (outer
monomers) and inactive (inner monomers)
conformation (22). We thus choose chains B
and D as possible templates for comparative
modeling. Since several residues are not
resolved in chain D, chain B was chosen as
the template for the A3C model.
Recently, two solution structures (PDB
identifiers 2JYW, 2KBO) and a crystal
structure (PDB identifier 3E1U, 2.3 Å
resolution) of A3G-CD have been obtained
(20, 21, 23). Superposition of A2 and the
crystal structure of A3G-CD (21) reveals the
common fold of the two polypeptides despite
their relatively low sequence identity (< 30%)
(Fig. 1). Notably, in the solution structure of
A3G-CD, one β-strand adopts a loop
conformation (20). As this segment is
anticipated to be involved in domain
dimerization in native full-length A3G, it has a
limited suitability as a template structure for
comparative modeling of A3C.
Stereochemical quality was higher in the A2
structure (Fig. S1). As we anticipate A3C to
form oligomers and A3G-CD was crystallized
as a monomer, while A2 has been
crystallized as a tetramer, oligomerization
properties might be better represented by
native A2 as a template than by A3G-CD.
Modeling of A3C on both templates at the
same time led to poor stereochemistry of the
resulting models, which could not be resolved
by geometry optimization (data not shown).
Alignments of A2 and A3G-CD to A3C were
carried out using MODELLER (24), explicitly
considering structural information of the
templates (Fig. S2). We yielded favorable
BLAST (25) e-values (A3C to A2: 6×10-23,
A3C to A3G-CD: 10-30; BLOSUM62). All
residues involved in the Zn2+ coordination are
conserved in both alignments. Insertions and
deletions were placed in loop regions.
Alignments are supported by matching
secondary structure predictions (PSIPRED
(26)) to those from the crystal structures
(data not shown). For both models, there is
no correspondence in the templates for the
N-terminal amino acids of A3C, so this region
had to be constructed without template.
Residue conservation was mapped back to
the templates (Fig. 1a, c) and is substantially
higher in the protein core and around the
Zn2+-coordinating center, showing a striking
pattern of alternating conserved / non-
conserved residues in buried / exposed parts
of both α-helices (conserved: i, i+3, ...) and β-
strands (conserved: i, i+2, ...).
Chapter I
18
Fig. 1. (A). Structure of A2, chain B. Conserved residues in A3C are shown in blue. Black sphere: catalytic
Zn2+-ion. (B) Superposition of A2 (white) and A3G, catalytic domain (A3G-CD, orange). Loop regions are not
shown. (C) Structure of A3G-CD. Residues conserved in A3C are shown in blue.
Ten initial models were built for each
template, energy-minimized and evaluated
for robustness (27). For each of the
templates, the model with the fewest
violations of the stereochemistry was
comparable to the quality of the template
structures (Fig. S1, Fig. S3) and subjected to
three independent runs of 20 ns molecular
dynamics (MD) simulations (Fig. S4).
Convergence of RMSD and energy
parameters suggested the fold to be stable
(Fig. S5). Identical protocols were applied for
the template structures, also showing
convergence. Simulated B-factors were
calculated from the trajectories for the
template structures as described (28),
averaged for each system and compared to
the experimental B-factors reported for the X-
Ray structures (r = 0.68 for A2, r = 0.76 for
A3G-CD, Fig. S5), suggesting the dynamics
of the structure to be well captured by our
MD simulation (29). Taken together, these
results suggest that both models of A3C are
valid from a structural point of view and thus
useful to deduce further hypotheses.
While the two minimized models show a
moderate pairwise RMSD of 2.7 Å (without
the N-terminal amino acids in the loop region
preceding the β-sheet), all residues
considered in this study are located at
overlapping positions (RMSD 0.7-1.4 Å) in
the two models within precision expected
from given levels of sequence identity and
thus are practically equivalent (Fig. 2). Here,
only the model structure of A3C based on the
structure of A2 is shown. All experiments
conducted in this study have been replicated
with the A3G-CD-based model without
significant change of predictions (data not
shown).
Fig. 2. Superposition of model of A3C, derived
from A2 (white) and A3G-CD (orange). Residues
shown to be of functional importance in this study
are shown in red.
A3C functions as a dimer. For A3C, which possesses only one domain,
it is reasonable to assume a mode of
dimerization analogous to that of A2, where
the β-strands of two monomers build a single
extended sheet. We would assume a similar
mode of domain interaction in full-length
A3G. We posed the questions whether i) A3C
oligomerizes, and ii) there is differential
Chapter I
19
Fig. 3. (A) Dimerization pose of A2 (experimental, white) and A3C (predicted by ClusPro, blue). Loop regions
are not shown. (B) Residues in the predicted dimerization interface of A3C (red) were mutated to alanine. (C) Immunoblot analysis of the expression and Vif-dependent degradation of wild-type(wt) A3C and different
dimerization mutants (K51A, F55A, W74A). A3C constructs were detected by an anti(α)-HA antibody. Tubulin
(Tub) served as loading control. (D) Antiviral activity of dimerization mutants F55A, W74A, K51A and wt A3C
against SIVagm, compared to non-transduced cells (no virus) and vector only control without A3C. wt or Δvif
SIVagmluc (VSV-G) virions were generated by co-transfection with the respective A3C mutant. Virions were
normalized by RT activity. Luciferase activity was measured 3 days post infection.
activity between the monomeric and
oligomeric forms.
First, protein-protein interaction interfaces
were predicted using ProMate (30). Amino
acids corresponding to the A2 dimerization
interface were highlighted as potentially
involved in the interaction. The protein
docking and clustering technique ClusPro
(31) accurately reproduced the A2 dimer
(RMSD = 1.7 Å), and was applied to
generate an A3C dimer model in silico (Fig.
3a). The overall topology of both the A2 and
A3C dimer models is similar.
Selected amino acids in the predicted
dimerization interface of A3C were mutated.
As exposed aromatic amino acids often
participate in protein-protein-interaction (32),
two prominent aromatic amino acids (F55,
W74) and K51, possibly contributing to
electrostatic interactions, were mutated to
alanine (Fig. 3b). All constructs (K51A, F55A,
W74A) were expressed, showing wt-like
degradation in presence of Vif (Fig. 3c). To
determine whether these A3C-mutants
display antiretroviral activity, virions were
generated by co-transfection with A3C
expression plasmids. In transduced cells,
A3C and K51A reduced the infectivity of the
Δvif SIV ~90-120-fold, F55A and W74A
showed a clearly diminished inhibitory activity
(~2-4-fold inhibition of Δvif viruses) (Fig. 3d).
In contrast, wt SIV was not inhibited by any of
the A3C-mutants. Inhibition of Δvif SIV by wt
A3C and K51A was shown to be dose-
dependent, whereas F55A and W74A were
still inactive although using highest amounts
of expression plasmid (Fig. 4).
Chapter I
20
Fig. 4. (A) Immunoblot analysis of the dose-dependent expression in 293T cells and encapsidation in viral
particles of wt A3C and different dimerization mutants (F55A, W74A and K51A). pcDNA, transfection control;
vector, wt SIVagm without A3C. The respective amounts of the A3C constructs were detected by an anti(α)-
HA antibody. Tubulin (Tub) and capsid (p27) served as loading control. (B) Dose-dependent antiviral activity
of dimerization mutants F55A, W74A and K51A against SIVagm Δvif, compared to wild-type(wt)-
huAPOBEC3C (A3C) and background of non-transduced cells (no virus) and vector only control (vector)
without A3C. Δvif SIVagm luc (VSV-G) virions were generated by co-transfection with the indicated amount
(µg) of the respective A3C mutant. Virions were normalized by RT activity and used for transduction.
Luciferase activity (in counts per second, cps) was measured 3 days post infection.
Packaging of A3 into the virion is crucial for
its antiretroviral activity (4, 8, 9). To
determine whether missing encapsidation of
the inactive mutants, is responsible for the
absence of inhibition, virions were generated
and analyzed for A3C content by immunoblot
analysis. Fig. 5a shows that all mutants were
efficiently packaged, comparable to wt A3C.
To determine whether oligomerization of the
mutant proteins correlates with antiretroviral
activity, expression vectors for V5-tagged wt
A3C were co-transfected with expression
plasmids for HA-tagged A3Cs (wt or
mutants). F55A and W74A barely
precipitated V5-tagged wt A3C. In contrast,
K51A and HA-tagged wt A3C were able to
precipitate V5-tagged wt A3C (Fig. 5b).
Quantification of the precipitated V5-tagged
wt A3C in presence of wt A3C compared to
W74A resulted in a significant higher binding
efficiency (~10 fold, Fig. S6). W74A-HA co-
precipitated the V5-tagged wt A3C only very
weakly. Crosslinking of wt A3C in total cell
lysate of transfected cells showed
monomeric, dimeric and tetrameric forms,
whereas W74A only formed monomers and
dimers (Fig. S6). We postulate that the W74A
mutation in the dimerization interface
prevents the formation of the inner dimer, but
does not influence the formation of the outer
dimer. This correlates with the observation of
a weaker binding activity seen in
immunoprecipitation assuming that the inner
dimer is more stable, where as the outer
Chapter I
21
Fig. 5. (A) Immunoblot analysis of A3C packaging.
293T cells were co-transfected with SIVagm Δvif
luc (VSV-G), the respective HA-tagged A3C
construct (mutants F55A, W74A, and wild-type
(wt)). Virions were harvested and normalized by
RT. Physically equal amounts of virions were
lysed and subjected to immunoblot analysis.
Presence of A3C in the virons was detected using
anti(α)-HA-antibodies. p27 (capsid) served as
loading control. (B) Immunoblot analysis of A3C
dimerization. 293T cells were co-transfected with
the respective HA-tagged A3C construct (mutants
K51A, F55A, W74A, and wt) and V5-tagged wt
A3C. Immune precipitates (IP) were subjected to
immunoblot analysis. Proteins were probed using
α-HA- / -V5-antibodies, respectively. Tubulin (Tub)
served as loading control in cell lysates.
dimer gets disrupted during the washing
steps. These results indicate that F55 and
W74 participate in dimerization of A3C, and
that oligomerization is crucial for antiviral
activity of the enzyme. It is noteworthy that
the dimer mutants did not exhibit DNA editing
activity in an E. coli mutation assay
(compared to wt A3C, Fig. 6), which is in
perfect agreement with the requirement for
dimeric protein for enzymatic activity.
A cavity of A3C mediates its
encapsidation.
Enzyme-substrate interactions typically occur
over well-defined protein binding pockets,
where the active site typically is one of the
largest cavities of the protein (33). Our
software PocketPicker (33) was employed to
identify potential ligand binding pockets in the
A3C model. The presumed active site was
found to be the 5th biggest cavity in the A2-
based structure (A3C/A2) (~80 Å3). The
largest cavity (~200 Å3) is found 15 Å apart
from the Zn2+ ion and has not yet been
described in literature (Fig. 7a). The A3G-
CD-based model structure of A3C contains a
similar binding pocket (Fig. 7b).
Fig. 6. Editing activity of wt and mutant A3C
proteins measured as RifR colonies per 109 viable
cells in E. coli mutation assay.
Four residues near this pocket were mutated
to alanine: K22, T92, R122, and N177 (Fig.
7c). Co-expression of these mutants with Vif
demonstrated protein degradation similar to
wt A3C (Fig. 8a). To test whether mutant
A3C-proteins inhibit virus replication, wt and
Δvif SIV were generated by co-transfection
with A3C expression plasmids. Wild-type
A3C as well as its mutants K22A, T92A, and
N177A inhibited Δvif SIV (~60-84-fold) but
not wt SIV (Fig. 8b). In contrast, the mutant
Chapter I
22
Fig. 7. (A) Model structure of A3C, derived from A2 (white). The binding pocket distal to the Zn2+-ion (black
sphere) is indicated in blue, R122 in red. (B) Model structure of A3C, derived from A3G-CD (orange). The
binding pocket is indicated in blue, the Zn2+-ion in black, R122 in red. (C) To test the function of this protein
cavity, residues in red were mutated to alanine.
R122A lost the inhibitory activity. Immunoblot
analysis of A3C content in viral particles
showed a greatly reduced packaging of
R122A compared to A3C wt and K22A,
T92A, N177A (Fig. 8c), although it was
detectable in cell lysates (Fig. 6a). By fusing
VPR to R122A the mutant could be re-
targeted into viral particles and showed
antiviral activity (Fig. 8d, e). In addition,
R122A showed DNA editing activity in
bacteria (Fig. 6). We conclude that R122 is
critically relevant for particle packaging but
not for antiviral activity of A3C.
In search for potential ligands of this
presumed binding pocket, it was compared to
pockets extracted from the PDBBind (34)
collection with known protein function and
ligands. The four hits that were identified as
most similar stem from human papillomavirus
(HPV) type 11 E2 transactivation domain
(TAD) complex (PDB identifier: 1R6N), a
DNA binding protein of HPV; and three holo-
structures of bovine RNase A (PDB
identifiers: 1QHC, 1JN4, 1O0M), co-
crystallized with different nucleotides. For the
A3C/A3G-CD pocket, four holo-structures of
bovine Rnase A (PDB identifiers: 1U1B,
1W4P, 1O0M, 1QHC) were among the eight
top scoring pockets. The natural substrates
of these pockets are nucleic acids: ds DNA
for HPV 11 E2 TAD, and ss and dsRNA for
bovine RNase A. Steric and electrostatic
properties of the hypothesized pocket in A3C
would allow for nucleic acids as a substrate,
possibly by R122 interacting with the
negatively charged sugar-phosphate
backbone.
To test for RNAs interacting with this binding
pocket, wt A3C and R122A were precipitated
from transfected cells, interacting RNA was
isolated and detected by 32P-labelling (Fig.
8f). Mutation of R122 resulted in strongly
decreased amounts of RNA bound to the
protein compared to wt A3C or a C98S active
site mutation. The isolated RNA was further
subjected to RT-PCR to amplify 7SL or 5.8S
RNA (Fig. 8g). A3C wt protein showed an
interaction with 7SL and 5.8S RNA, while the
mutant R122A lost the binding to 5.8S RNA.
Furthermore, RNA binding was shown to be
crucial for interaction of A3C to SIV-
nucleocapsid (NC). A3C wt interacts in a
RNA-dependent manner with SIV-NC, while
R122A exhibits no binding activity (Fig. 9).
Chapter I
23
Fig. 8. (A) Immunoblot analysis of the expression and Vif-dependent degradation of wild-type(wt) A3C and the
mutants K22A, T92A, R122A, N177A. The respective A3C constructs were detected by an anti(α)-HA
antibody. Tubulin (Tub) served as loading control. (B) Antiviral activity of the mutants K22A, T92A, R122A and
N177A against SIVagm, compared to wt A3C and background of non-transduced cells (no virus) and vector
only control (vector) without A3C. 293T cells were co-transfected with wt or Δvif SIVagmluc (VSV-G),
respectively, and the respective A3C mutant. Virions were normalized by RT and HOS cells transduced.
Luciferase activity was determined at 3 days post infection. (C) Immunoblot analysis of A3C packaging. 293T
cells were co-transfected with Δvif SIVagmluc(VSV-G), the respective HA-tagged A3C construct (mutant R122A
and wt). Virions were harvested and normalized by RT. Physically equal amounts of virions were lysed and
subjected to immunoblot analysis. Presence of A3C in the virons was detected using α-HA-antibodies. p27
(capsid) served as loading control. (D) Antiviral activity of Vpr-A3C and Vpr-R122A fusion proteins against
SIVagm, compared to wt A3C and R122A and background of non-transduced cells (no virus) and vector only
control without A3C. Δvif SIVagmluc (VSV-G) virions were generated by co-transfection with the respective A3C
mutant. Virions were normalized by RT activity and used for transduction. Luciferase activity was measured 3
days post infection. (E) Immunoblot analysis of the expression and encapsidation of wt A3C and R122A
compared to the respective Vpr-fusion-proteins. A3C constructs (+ or – Vpr) were detected by an anti(α)-HA
antibody. Tubulin (Tub) served as loading control for cell lysates and p27 (capsid) for viral lysates. (F) RNA
interacting with A3 proteins. A3C wt or mutant proteins and wt A3G were expressed in 293T and cell lysates
were subjected to immuno precipitation. RNA bound to immunoprecipitated proteins was radioactively labeled
with 32P through RT-PCR and separated on a 12% PAA Gel and exposed on X-ray film. Background was set
to signal of untransfected cells (mock). Equal amount of precipitated A3 protein was proven by immunoblot
analysis of the elution fraction with an anti(α)-HA antibody. (G) RT-PCR on RNA interacting with A3 proteins.
Isolated and A3 bound RNA (IP) was reverse transcribed and amplified using specific primers for 7 SL and 5.8
S RNA. Background signal was determined with RNA from untransfected cells (mock). Availability of the
tested RNAs was confirmed for each sample through RT-PCR on RNA from cells prior to IP (cells).
We conclude that the large pocket detected
on the surface of A3C plays a key role in
incorporation of A3C into viral particles,
mediated by RNA-dependent interaction with
SIV-NC.
Chapter I
24
DISCUSSION
We have presented two three-dimensional
models of A3C derived by comparative
protein structure modeling taking the crystal
structures of A2 and the catalytic domain of
A3G as templates. These models were used
to deduce hypotheses regarding dimerization
and to characterize a presumed substrate
binding pocket of A3C. Although sequence
identity between A2 and A3C falls into the
“twilight zone” (35, 36), homology between
A2 and A3C can be assumed due to the
conservation of the Zn2+-coordinating
domain, the comparable class of enzyme
function, and predicted similar secondary
structure. The stereochemical quality of
energy-minimized models of A3C was
comparable to that of the templates, and
folding stability was demonstrated by MD
simulations. The level of sequence identity of
the templates to the targets a priori indicates
an expected medium accuracy of the model
(~85% of residues within 3.5 Å of the actual
conformation), rendering them suitable to
support site-directed mutagenesis
experiments (37), although predictions
requiring exact side chain orientations cannot
be made.
Fig. 9. Immunoblot analysis of A3C interaction with nucleocapsid (NC) from SIVagm. 293T cells were
transfected with A3C wt or R122A expression plasmids and incubated with GST or GST-NC expressed in E.
coli. After RNaseA treatment and immunoprecipitation, bound proteins were subjected to immunoblot analysis.
Proteins were detected using anti-HA mAb or anti-GST mAb, respectively. Equal expression of wt A3C and
R122A in 293T cells was confirmed (input).
Using an automated approach we suggest a
potential substrate binding pocket in A3C,
which is distal from the Zn2+-coordinating
site. Mutation of R122 at the pocket entrance
resulted in the loss of antiviral activity due to
diminished incorporation of the mutated
protein into the virion. This arginine is
conserved in A2 A3G-CD and A3G-ND (Fig.
S2). Packaging of A3G has been
demonstrated to be dependent on interaction
with 7SL RNA (38). As a mutation of R122 in
A3C impedes the RNA-dependent NC-
interaction and the incorporation into virions
one can speculate on R122 being important
for RNA dependent packaging of the protein.
With regard to the accuracy level of our
protein model, it is possible that this binding
pocket partially overlaps with the active site,
resulting in a large binding pocket of bipartite
function, similar to the substrate binding
channel proposed for A3G-CD (21). R122
could then be thought of interacting with DNA
as a substrate anchor for deaminase activity.
Both activities might be mutually
independent, as suggested by comparing the
Chapter I
25
characteristics of active-site mutant C98S
with R122A.
Our A3C model is consistent with dimer
formation of A3C analogous to that of A2.
Mutations in the predicted interaction surface
revealed that the antiviral function of A3C
requires dimerization. In contrast to previous
data (39), it was later shown that monomeric
A3G is an active inhibitor of Δvif HIV (40).
Because of the inherent dimeric character of
A3G, which possesses two copies of the
Zn2+-coordinating motif, additional
dimerization of A3G might not be required for
antiviral activity.
Summarizing, our results demonstrate that a
predicted binding pocket of A3C interacts
with RNA, e.g. 5.8 S RNA, and that RNA
interaction is required for encapsidation
mediated by binding to NC, but not for
antiviral activity. Why dimerization of A3C is
critical for antiviral activity remains an open
question and an important subject for future
studies.
During revision of this manuscript, Huthoff et
al. (41) presented a homology model of A3G
and demonstrated its RNA-dependent
packaging that is determined by a residue
inside its N-terminal domain being equivalent
to R122 (for sequence alignment see Fig.
S2), thereby additionally supporting our
hypothesis.
EXPERIMENTAL PROCEDURES
Model building. Target A3C sequence, NCBI
accession: NP_055323. Chain B of the tetrameric
crystal structure of A2 (PDB identifier 2NYT, 2.5 Å
resolution (22, 42)) and catalytic domain of A3G
(PDB 3E1U, 2.3 Å resolution (21)) served as
template. The initial sequence-alignment of A3C to
monomeric A2, chain B, and the catalytic domain
of A3G, was performed by the align2d function of
MODELLER 9v4 (24, 43). Both target-template
alignment and structural coordinates of A2, chain
B, and A3G-CD, were used to build ten initial
models by satisfaction of spatial restraints,
subjected to energy minimization and MD
simulation (Supplementary Material).
Characterization of binding pockets. PocketPicker (33) was used to automatically
identify potential binding pockets and encode
them as correlation vectors as described (33, 44).
Each vector was then compared to 1,300 binding
pockets with annotated function from the “refined
set” of the PDBBind data set (34) using the
Euclidean distance metric.
Mapping of functional residues. MOE 2006.08
(Chemical Computing Group, Montreal, Canada)
was used to calculate the solvent accessible
surface of the protein model and map electrostatic
properties by a Poisson-Boltzmann potential.
Putative protein-protein interaction interfaces were
selected manually by looking for solvent-exposed
hydrophobic patches, and fully automated using
ProMate (30). Protein-protein docking and
clustering of docking poses was carried out with
ClusPro (31).
Plasmids. C-terminally hemagglutinin (HA)-
tagged A3C expression plasmid has been
described (45). Viral vectors were produced by co-
transfecting pSIVagm Δvif luc (4), pMD.G, a VSV.G
expression plasmid, and supplemented by pcVif-
SIVagm-V5 (46). For co-immunoprecipitation
studies pcDNA3.1-APOBEC3C-V5-6xHis (47) was
used. HA-tagged mutated A3C constructs were
derived by fusion PCR and cloned into pcDNA
3.1(+) (Invitrogen), using BamHI and NotI
restriction sites. The pcVPR-A3C-HA expression
plasmid was generated by fusion of SIVagm-VPR
cDNA to the N-Terminus of HA-tagged wt or
mutant A3C with a Gly4-Ser-linker and cloned into
pcDNA 3.1(+) using the same restriction sites. The
premature stop-codon in VPR of SIVagm_TAN-1
was rectified by site directed mutagenesis.
Sequences of primers in Supplementary Material.
PCRs were performed with Phusion™ DNA
Polymerase (Finnzymes) at: 1 cycle at 98°C, 3
Chapter I
26
min; 35 cycles at 98°C, 15 s; 65-71°C, 30 s; and
72°C, 20 s; and 1 cycle at 72°C, 10 min.
Immunoblot analysis. For analysis of expression
of A3C-constructs, 293T cells were transfected
with 1 µg A3C-expression plasmid and 2 µg of Vif-
expression plasmid, using LipofectamineLTX
(Invitrogen). Two days post transfection, cells
were harvested and lysed using Western Lysis
Buffer (100 mM NaCl, 20 mM Tris, pH 7.5, 10 mM
EDTA, 1% sodium deoxycholate, 1% Triton X-100,
and complete protease inhibitor (Roche)). Lysates
were cleared by centrifugation and subjected to
SDS-PAGE followed by transfer to a PVDF
membrane. A3C-HA-constructs were detected
using an anti (α)-HA antibody (1:104 dilution,
Covance) and α-mouse horseradish peroxidase
(1:7,500 dilution, Amersham Biosciences). For
detection of Vif-V5 or A3C-V5, an α-V5 antibody
(1:4,000 dilution, Serotec) was applied. α -tubulin
was detected using an α-tubulin antibody (1: 104
dilution, Sigma). Signals were visualized by ECL
plus (Amersham Biosciences).
Packaging of A3C. To detect A3C in virions, A3C
expression constructs were co-transfected with
pSIVagm Δvif luc and pMD.G in 293T cells, as
described above. Particles were precipitated by
ultracentrifugation over a 20% sucrose cushion,
normalized for activity of reverse transcriptase and
lysed using Western Lysis Buffer. Lysates were
directly subjected to SDS-PAGE and transferred
to a PVDF-membrane. p27 was detected using a
p24/p27 monoclonal antibody AG3.0 (1:250) (48).
Co-immunoprecipitation. To detect protein
interaction, expression plasmids of the respective
proteins were co-transfected into 293T-cells. Cells
were lysed in ice cold lysis buffer (25mM Tris (pH
8.0), 137mM NaCl, 1% Glycerol, 0.1% SDS, 0.5%
Na-deoxycholat, 1% NP-40, 2mM EDTA and
complete protease inhibitor cocktail (Roche)). The
cleared lysates were incubated with 30 µL α-HA
Affinity Matrix Beads (Roche), 60 min, 4°C. The
samples were washed 5 times with ice cold lysis
buffer. Bound proteins were eluted by boiling the
beads for 5 min at 95°C in SDS loading buffer.
Immunoblot analysis and detection was done as
described. Light units of the elution fractions were
directly quantified from membranes incubated with
ECL plus using the Lumianalyst 3.0 software
(Roche).
Reporter virus assay. To measure antiretroviral
activity of A3C-constructs, expression plasmids
were co-transfected to 293T cells with pSIVagmΔvif
luc, pMD.G, in presence and absence of a Vif
SIVagm expression plasmid. Two days after
transfection, supernatants were harvested,
normalized by RT activity and transduced 2 × 103
HOS cells in a 96-well dish. Three days after
transduction, intracellular luciferase activity was
quantified using Steady Lite HTS (Perkin Elmer).
Data are presented as the average counts/second
of the triplicates ± standard deviation. RT activity
was quantified using the Lenti-RT Activity Assay
(Cavidi Tech). The cell lines 293T and HOS were
cultured in Dulbecco’s high glucose modified
Eagle’s medium (Invitrogen), 10% fetal bovine
serum, 0.29 mg/ml L-glutamine, and 100 units/m
penicillin/streptomycin at 37°C with 5% CO2.
A3C-RNA-interaction. To detect protein-RNA
interaction, expression plasmids of the respective
proteins were transfected to 293T-cells as
described above. Cells were lysed in ice cold lysis
buffer (PBS with 1% Triton X-100, 16U/mL
RiboLock RNase Inhibitor (Fermentas) and
complete protease inhibitor cocktail (Roche)). The
cleared lysates were incubated with 50 µL α-HA
Affinity Matrix Beads (Roche), 60 min, 4°C. The
samples were washed 5 times with ice cold lysis
buffer. RNA bound to immobilized proteins was
extracted from HA-beads using TRIzol reagent
(Invitrogen) (according to the companies
instructions). RNAs co-precipitated with A3C
proteins were dissolved in DEPC-treated H2O and
equal amounts were used for reverse transcription
using random hexamer primers (Fermentas) in the
presence of α-32P-dATP (Hartmann Analytic).
Radioactive labeled cDNA was separated on a
12% TBE 8M Urea PAA Gel and visualized with
Amersham Hyperfilm (GE healthcare). Specific
RT-PCR on A3C bound RNAs was performed
using Reverd Aid First Strand cDNA synthesis kit
(Fermentas) and random hexamer primers.
Chapter I
27
Protein-Protein-Crosslinking. For crosslinking
experiments, expression plasmids of the
respective proteins were transfected to 293T-cells,
two days later cells were lysed and whole cell
lysate was incubated with 50mM N-Ethylmaleimid
(NEM/Calbiochem) for 2hrs at RT. Immunoblot
analysis was performed as described above but
no DTT or β-mercapthoethanol was added to the
SDS-loading buffer during electrophoresis.
E. coli mutation assay Uracil DNA glycosylase-deficient E. coli strain
BW310 was transformed with isopropyl 1-thio-β-D-
galactopyranoside (IPTG) - inducible APOBEC3
expression constructs (APOBEC3 encoding cDNA
was inserted in pTrc99A using XhoI and SalI
restriction sites) or pTrc99A empty vector.
Individual colonies were picked and grown to
saturation in a rich medium containing 100µg/ml
ampicillin and 1 mM IPTG. Appropriate dilutions
were plated onto a rich medium containing
100µg/ml rifampicin to select RifR colonies after an
overnight incubation. Mutation frequencies were
reported as the number of RifR colonies per 109
viable cells.
A3C – NC interaction
Plasmids: A bacterial plasmid (pGEX-p2NCp1-
SIVtan) expressing p3:NC:p1 fused to GST on its
N-terminus was generated by PCR amplification of
relevant part of SIVagmTAN and subsequent
insertion into pGEX-6P via XmaI and XhoI
restriction sites.
Escherichia coli (strain: BL21 DE3 Rosetta)
transformed with pGEX-p2NCp1-SIVtan and
parental pGEX-6p-1 were used for protein
expression. Cells were harvested 5 h post
induction with 100 µM IPTG (Fermentas), lysed by
sonification in PBS supplemented with 1 % NP 40
and protease inhibitor coctail (Calbiochem).
Glutathione sepharose beads (GE Healthcare)
were incubated with clarified lysates for 30 min at
25 °C, washed 3 times with lysis buffer (10fold
beads volume), resuspended in 10 mM Tris pH
7,5; 150 mM NaCl; 0,5 % NP40 with or without 34
µg/ml RNaseA (Fermentas) and incubated for 1h
at 37 °C. Subsequently samples were washed and
incubated with lysates of 293T cells expressing wt
A3C or R122A mutant over night at 4 °C in PBS,
1% NP 40 and 300 mM NaCl. After the beads
were washed extensively with PBS/1 % NP 40,
proteins were eluted with 10 mM glutathione, 50
mM TRIS (pH 8) and used for immunoblot
analysis using anti-HA mAb or anti-GST mAb
(1:1.000 dilution; cell signaling).
ACKNOWLEDGMENTS
We thank Marion Battenberg, Elea Conrad and
Norbert Dichter for expert technical assistance,
and Nathaniel R. Landau and Bryan Cullen for the
gift of reagents. The following reagents were
obtained through the NIH AIDS Research and
Reference Reagent Program, Division of AIDS,
NIAID, NIH: pcDNA3.1-APOBEC3C-V5-6XHis
from B. Matija Peterlin and Yong-Hui Zheng,
monoclonal antibody to HIV-1 p24 (AG3.0) from
Jonathan Allan. We are grateful to the Chemical
Computing Group for providing an MOE license.
This study was supported by the Beilstein-Institut
zur Förderung der Chemischen Wissenschaften,
and the Deutsche Forschungsgemeinschaft (SFB
579, project A11.2). Part of the study was funded
by DFG grant 1608/3-1 to CM. CM is supported by
the Ansmann foundation for AIDS research. We
thank Dieter Häussinger for support. MW and GS
are grateful to Boehringer-Ingelheim Pharma for
funding.
References
1. Sheehy AM, Gaddis NC, Choi JD &
Malim MH (2002) Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein. Nature 418, 646-650.
2. Lecossier D, Bouchonnet F, Clavel F & Hance AJ (2003) Hypermutation of HIV-1 DNA in the absence of the Vif protein. Science 300, 1112.
Chapter I
28
3. Mangeat B et al. (2003) Broad antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. Nature 424, 99-103.
4. Mariani R et al. (2003) Species-specific exclusion of APOBEC3G from HIV-1 virions by Vif. Cell 114, 21-31.
5. Zhang H et al. (2003) The cytidine deaminase CEM15 induces hypermutation in newly synthesized HIV-1 DNA. Nature 424, 94-98.
6. Bishop KN et al. (2004) Cytidine deamination of retroviral DNA by diverse APOBEC proteins. Curr. Biol. 14, 1392-1396.
7. Holmes RK, Malim MH & Bishop KN (2007) APOBEC-mediated viral restriction: not simply editing? Trends Biochem. Sci. 32, 118-128.
8. Marin M, Rose KM, Kozak SL & Kabat D (2003) HIV-1 Vif protein binds the editing enzyme APOBEC3G and induces its degradation. Nat. Med. 9, 1398-1403.
9. Sheehy AM, Gaddis NC & Malim MH (2003) The antiretroviral enzyme APOBEC3G is degraded by the proteasome in response to HIV-1 Vif. Nat. Med. 9, 1404-1407.
10. LaRue RS et al. (2009) Guidelines for naming nonprimate APOBEC3 genes and proteins. J. Virol. 83, 494-497.
11. Münk C et al. (2008) Functions, structure, and read-through alternative splicing of feline APOBEC3 genes. Genome Biol. 9, R48.
12. Jonsson SR et al. (2006) Evolutionarily conserved and non-conserved retrovirus restriction activities of artiodactyl APOBEC3F proteins. Nucleic Acids Res. 34, 5683-5694.
13. Jarmuz A et al. (2002) An anthropoid-specific locus of orphan C to U RNA-editing enzymes on chromosome 22. Genomics 79, 285-296.
14. Chen H et al. (2006) APOBEC3A is a potent inhibitor of adeno-associated virus and retrotransposons. Curr. Biol. 16, 480-485.
15. Harari A, Ooms M, Mulder LC & Simon V (2009) Polymorphisms and splice variants influence the antiretroviral activity of human APOBEC3H. J. Virol. 83, 295-303.
16. OhAinle M, Kerns JA, Malik HS & Emerman M (2006) Adaptive evolution and antiviral activity of the conserved mammalian cytidine deaminase APOBEC3H. J. Virol. 80, 3853-3862.
17. Langlois MA, Beale RC, Conticello SG & Neuberger MS (2005) Mutational comparison of the single-domained APOBEC3C and double-domained APOBEC3F/G anti-retroviral cytidine deaminases provides insight into their DNA target site specificities. Nucleic Acids Res. 33, 1913-1923.
18. Yu Q et al. (2004) APOBEC3B and APOBEC3C are potent inhibitors of simian immunodeficiency virus replication. J. Biol. Chem. 279, 53379-53386.
19. Bourara K, Liegler TJ & Grant RM (2007) Target cell APOBEC3C can induce limited G-to-A mutation in HIV-1. PLoS. Pathog. 3, 1477-1485.
20. Chen KM et al. (2008) Structure of the DNA deaminase domain of the HIV-1 restriction factor APOBEC3G. Nature 452, 116-119.
21. Holden LG et al. (2008) Crystal structure of the anti-viral APOBEC3G catalytic domain and functional implications. Nature 456, 121-124.
22. Prochnow C et al. (2007) The APOBEC-2 crystal structure and functional implications for the deaminase AID. Nature 445, 447-451.
23. Furukawa A et al. (2009) Structure, interaction and real-time monitoring of the enzymatic reaction of wild-type APOBEC3G. EMBO J.
24. Eswar N et al. (2007) Comparative protein structure modeling using MODELLER. Curr. Protoc. Protein Sci. Chapter 2, Unit.
25. Altschul SF et al. (1990) Basic local alignment search tool. J. Mol. Biol. 215, 403-410.
26. Jones DT (1999) Protein secondary structure prediction based on position-specific scoring matrices. J. Mol. Biol. 292, 195-202.
27. Melo F & Feytmans E (1998) Assessing protein structures with a non-local atomic interaction energy. J. Mol. Biol. 277, 1141-1152.
Chapter I
29
28. Hunenberger PH, Mark AE & van Gunsteren WF (1995) Fluctuation and cross-correlation analysis of protein motions observed in nanosecond molecular dynamics simulations. J. Mol. Biol. 252, 492-503.
29. Bond PJ, Faraldo-Gomez JD, Deol SS & Sansom MS (2006) Membrane protein dynamics and detergent interactions within a crystal: a simulation study of OmpA. Proc. Natl. Acad. Sci. U. S. A 103, 9518-9523.
30. Neuvirth H, Raz R & Schreiber G (2004) ProMate: a structure based prediction program to identify the location of protein-protein binding sites. J. Mol. Biol. 338, 181-199.
31. Comeau SR, Gatchell DW, Vajda S & Camacho CJ (2004) ClusPro: a fully automated algorithm for protein-protein docking. Nucleic Acids Res. 32, W96-W99.
32. Mitchell JBO et al. (1993) Amino/aromatic interactions. Nature 366, 413.
33. Weisel M, Proschak E & Schneider G (2007) PocketPicker: analysis of ligand binding-sites with shape descriptors. Chem. Cent. J. 1, 7.
34. Wang R, Fang X, Lu Y & Wang S (2004) The PDBbind database: collection of binding affinities for protein-ligand complexes with known three-dimensional structures. J. Med. Chem. 47, 2977-2980.
35. Abagyan RA & Batalov S (1997) Do aligned sequences share the same fold? J. Mol. Biol. 273, 355-368.
36. Rost B (1999) Twilight zone of protein sequence alignments. Protein Eng 12, 85-94.
37. Marti-Renom MA et al. (2000) Comparative protein structure modeling of genes and genomes. Annu. Rev. Biophys. Biomol. Struct. 29, 291-325.
38. Wang T et al. (2007) 7SL RNA mediates virion packaging of the antiviral cytidine deaminase APOBEC3G. J. Virol. 81, 13112-13124.
39. Shindo K et al. (2003) The enzymatic activity of CEM15/Apobec-3G is essential for the regulation of the infectivity of HIV-1 virion but not a sole determinant of its antiviral activity. J. Biol. Chem. 278, 44412-44416.
40. Opi S et al. (2006) Monomeric APOBEC3G is catalytically active and has antiviral activity. J. Virol. 80, 4673-4682.
41. Huthoff H et al. (2009) RNA-dependent oligomerization of APOBEC3G is required for restriction of HIV-1. PLoS. Pathog. 5, e1000330.
42. Berman HM et al. (2000) The Protein Data Bank. Nucleic Acids Res. 28, 235-242.
43. Sali A & Blundell TL (1993) Comparative protein modelling by satisfaction of spatial restraints. J. Mol. Biol. 234, 779-815.
44. Stahl M, Bur D & Schneider G (1999) Mapping of proteinase active sites by projection of surface-derived correlation vectors. J. Comput. Chem. 20, 336-347.
45. Muckenfuss H et al. (2006) APOBEC3 proteins inhibit human LINE-1 retrotransposition. J. Biol. Chem. 281, 22161-22172.
46. Perkovic M et al. (2008) Species-specific inhibition of APOBEC3C by the prototype foamy virus protein Bet. J. Biol. Chem.
47. Zheng YH et al. (2004) Human APOBEC3F is another host factor that blocks human immunodeficiency virus type 1 replication. J. Virol. 78, 6073-6076.
48. Simm M et al. (1995) Aberrant Gag protein composition of a human immunodeficiency virus type 1 vif mutant produced in primary lymphocytes. J. Virol. 69, 4582-4586.
Chapter II
30
CHAPTER II
VPR(HIV-1) fused to APOBEC3C alters its restriction properties but
not its sub-viral localization
Henning Hofmann and Carsten Münk
Heinrich-Heine-University Düsseldorf, Clinic for Gastroenterology, Hepatology and Infectiology, Building
23.12_U1 Room 81, Moorenstr. 5, 40225 Düsseldorf, Germany
Data not published. ABSTRACT
The APOBEC3 family of cytidine deaminases
protects vertebrates from viral pathogens.
Human APOBEC3C (A3C) restricts the
Simian immunodeficiency virus (SIV) in
absence of its Vif protein, but has no antiviral
effect against vif-deficient Human
immunodeficiency virus (HIV Δvif).
Our study confirmed that A3C lacks the
ability to restrict HIV Δvif although it is
incorporated into viral particles. This
phenotype could be altered through fusing
the Viral Protein R (Vpr) of HIV-1 either to the
C- or N- terminus of A3C. Interestingly, also
wt HIV was inhibited by the Vpr-A3C fusion
protein indicating a block of A3C interaction
with Vif. Exploring the sub-viral localization of
the fusion protein compared to wt A3C did
not show significant differences. Thus, HIV
likely encodes for a factor which counteracts
A3C activity additional to vif. Preliminary
experiments indicate that the viral integrase
(IN) acts as an A3C inhibitor.
INTRODUCTION
Although the APOBEC3 protein family is very
homologues sharing a characteristic motif for
Zn2+-coordination, the specificity of each
APOBEC3 differs from the others (1). Several
APOBEC3 proteins (e.g. A3F and A3G)
restrict HIV replication in the absence of the
Viral infectivity factor (Vif) protein (2, 3).
Another member, A3C is only active against
SIV and Long Interspersed Element-1 (LINE-
1) retrotransposons, but not against HIV (4,
5). The inhibitory effect of A3C against SIV
Δvif was shown to be 10-100 fold. However,
one report described an influence of A3C on
HIV in target cells (6). Here single G-A
mutations could be observed in spreading
HIV infection of cells expressing A3C. The
authors speculate on a potential benefit of
these mutations to the mutation rate of the
HIV genome. The viral counterpart that
mediates the HIV Δvif resistance against A3C
restriction was not described yet.
Another protein of the APOBEC3 family,
A3A, also has no antiviral activity against
HIV Δvif. This Vif-independent counteraction
of HIV could be circumvented by fusing the
Viral Protein R (Vpr) of HIV-1 to A3A (7).
Several functions in HIV replication have
been suggested for Vpr such as modest
transactivation of viral transcription (8),
Chapter II
31
contribution to the nuclear import of the pre-
integration complex (9, 10) or induction of a
cell cycle arrest in the G2/M phase in
mammalian cells (11, 12). Another
characteristic of Vpr is the interaction with the
p6 domain of the Gag precursor polyprotein
of HIV (13, 14). Via this interaction Vpr gets
efficiently incorporated into HIV particles and
enables the protein to deliver heterologous
proteins (as Vpr fusion proteins) into the virus
(15, 16). The fusion of Vpr to A3A resulted in
alterations of the sub-viral localization of the
fusion protein compared to wt A3A. The
inactive A3A was not localized within the viral
core whereas the antiviral active Vpr-A3A
fusion protein was found to be part of the
viral core thus being in close proximity to its
potential target, the viral genome. Further,
the fusion of Vpr of SIV of the African Green
Monkey (SIVagm) to a packaging deficient
R122A mutant of A3C (Vpragm.R122A) could
rescue its incorporation (see Chapter I).
The aim of this study was to identify whether
the fusion of HIV-1Vpr to A3C modulates its
restriction properties against HIV and / or
alters the localization of A3C within the viral
particle.
Fig. 1: A3C is active against SIV∆vif, but not against HIV∆vif. (A) Antiviral activity of A3C and A3G proteins
against SIVagm Δvif luc and HIV Δvif luc in presence (white bars) or absence (black bars) of the respective co-
expressed Vif protein compared to controls without A3 expression (HIV / SIV only). (B) Immunoblot analysis of
the expression, Vif-dependent degradation in producer cells and incorporation into virions of HA-tagged A3C
and A3G using an α-HA antibody. Co-expression of the respective Vif proteins of HIV and SIV is indicated with
a plus sign (+), the absence of Vif indicates a minus sign (-). Tubulin (α-Tub) and capsid (α-CA) served as
loading control.
Chapter II
32
RESULTS
A3C acts antiviral only against SIV∆vif
APOBEC3G (A3G) is well described as a
potent inhibitor of HIV ∆vif and SIV as well as
other viruses. The antiretroviral activity of
A3C is restricted to SIV ∆vif (5). A vif-
deficient HIV is not inhibited by A3C. As
incorporation into viral particles during
budding is crucial for the antiviral activity of
APOBEC3 cytidine deaminases, we first
examined whether the lack of restriction
against HIV ∆vif is due to the exclusion of
A3C from HIV particles (Fig.1). Therefore
virions were produced in 293T cells in
presence and absence of A3C and A3G as a
positive control. As expected A3G restricted
SIV in a vif-independent manner and HIV ∆vif
was also inhibited up to 30-fold (Fig.1A).
Immunoblot analysis of lysates from
transfected cells showed comparable
expression levels of A3C and A3G and a Vif-
dependent degradation of both proteins (Fig.
1B). As expected, A3G was not degraded by
Fig. 2: Schematic representation of the used
APOBEC3 and A3C-Vpr fusion proteins. APOBEC
proteins are shown in grey having one (A3C) or
two (A3G) Zn2+ - coordinating motifs shown in red.
HIV-1Vpr (blue) fused to the N- or C- terminus of
A3C with a Gly4-Ser- linker in between.
the Vif protein of SIVagm (17). In both cases
of viral restriction A3G was incorporated into
the viral particles (Fig.1B). HIV ∆vif was
resistant against A3C, but unexpectedly it
was also incorporated into viral particles. The
A3C-mediated restriction of SIV ∆vif
demonstrated that the deaminase was
enzymatically active (Fig.1A).
Fig. 3: Antiviral activity of wt A3C and indicated Vpr-A3C fusion proteins against HIV in presence (white bars)
or absence (black bars) of Vif fromHIV compared to non-transduced cells (mock) and vector only control
without A3C. Expression of the different HA-tagged proteins in producer cells was determined via immunoblot
analysis with an α-HA antibody. Tubulin (α-Tub) served as loading control.
Chapter II
33
A VPR(HIV-1) - A3C fusion protein exhibits
anti-HIV activity
Vpr of HIV was fused to the N- and the C-
terminus of A3C (Fig. 2) via fusion PCR. In
both cases a Gly4-Ser-linker was introduced
to separate the different protein domains.
The same vpr construct was also fused to the
N-terminus of the A3C C98S mutant. C98 is
part of the Zn2+-coordinating motif. Mutating
only one of the 4 amino acids that
coordinates the Zn2+-ion in the active site of
the protein results in loss of function proteins
(see Chapter III). Here, the Vpr.C98S
construct served as a negative control. To
investigate the antiviral potential of the Vpr
fusion proteins, HIV ∆vif particles were
produced in 293T cells in presence or
absence of the different expression plasmids.
Using luciferase reporter viruses, A3C
showed no antiviral activity against HIV ∆vif,
but both A3C Vpr fusion proteins reduced the
infectivity of HIV ∆vif ~90-fold. (Fig. 3A).
Interestingly, this effect was Vif-independent
because co-expression of HIV Vif could not
counteract the activity of Vpr.3C. This effect
was shown to be specific, as titration
experiments resulted in a dose-dependent
inhibition of ∆vif and wt HIV, whereas A3C
was inactive against HIV ∆vif at any
concentration of used expression plasmid
(Fig. 4). Fusion of Vpr to the C-terminus of
A3C resulted in an even stronger inhibition of
HIV (up to ~150-fold) in presence or absence
of Vif (Fig. 3A). The increased antiviral
activity of 3C.Vpr correlated with higher
expression compared to Vpr.3C (Fig. 3B).
The antiviral activity of 3C.Vpr was likely
caused by an enzymatic activity of the A3C
part of the fusion protein because fusion of
the enzymatic inactive C98S mutant to Vpr
did not restrict HIV (Fig. 3A).
Fig. 4: (A) Dose-dependent antiviral activity of Vpr.3C against HIVΔvif in presence (white bars) and absence
(black bars) of co-expressed HIV Vif, compared to wt A3C (grey bars) and vector only control without A3C. (B)
Immunoblot analysis of the dose-dependent expression of wt A3C and Vpr.3C in producer cells. The
respective amounts of A3C were detected by an α-HA antibody, tubulin (Tub) served as loading control.
Chapter II
34
VPR-fusion does not change sub-viral
localization
To explore the relevance of sub-viral
localization of A3C and Vpr.3C, VSV-G
pseudotyped vif-deficient HIV particles were
produced in the presence of A3C or Vpr.3C.
After concentration through a 20% sucrose
cushion the virions were applied to a 20-60%
OptiPrep™gradient (Fig. 5).
Fig. 5: Schematic representation of the
constitution of the 20 - 60% OptiPrep™ gradient.
Numbers (in %) indicate the OptiPrep™
concentration (diluted with H2O) in the different
compartments. TX-100 = Triton X–100.
A 1% TritonX-100 containing layer on top of
the gradient was used to destroy the
envelope structures resulting in a separation
of the viral core pelleting at fractions with a
higher density and the free proteins that can
be found in fractions with lower density (18-
20). After ultracentrifugation 1mL fractions
were collected from bottom to top and
analyzed by immunoblotting (Fig. 6B). Both,
A3C and Vpr.3C localized in the fractions 5 to
8 with a density of 1.5 to 2 g/mL (Fig 6A and
B). The reverse transcriptase (RT) is part of
the viral core and therefore can be used to
identify the core containing fractions of the
gradient. Measuring the RT-activity for all
fractions resulted in two peaks (Fig. 6A). The
first peak of RT in both gradients was found
in fractions co-localizing with A3C and
Vpr.3C. RT activity was also detectable in the
fractions on top of the gradient likely
indicating free RT, possibly as result of a too
harsh treatment with 1% TritonX-100
resulting in disruption of some viral cores.
The envelope protein (here VSV-
glycoprotein) was only seen in the top
fractions of the gradient supporting a proper
separation of free viral proteins from core
structures (Fig. 6B).
As A3C and Vpr.3C show identical sub-viral
localizations, modes of encapsidation cannot
explain the different restriction phenotypes of
both proteins. It is more likely that HIV is able
to circumvent the restriction of A3C by a virus
encoded factor.
HIV integrase as promising antagonist of
A3C mediated restriction
To investigate whether other HIV encoded
factors besides the Vif protein might be able
to counteract A3C, various viral proteins were
analyzed on their potential to interact with
A3C (data not shown). The HIV integrase
(HIV-IN) turned out to be the most promising
candidate. Integrase proteins of HIV-1 and
SIVmac with a C-terminal 6x Histidin tag
(His-tag) were expressed in E. coli and
bound to Ni-NTA agarose (Fig. 7, upper
panel). A3C was expressed in 293T cells and
supplied to the Ni-NTA bound integrases.
The HIV-IN showed interaction with A3C
whereas the integrase of SIV was not able to
bind A3C with the same efficiency (Fig. 7,
lower panel). To test the binding of IN to
Vpr.3C, both proteins, A3C and Vpr.3C, were
expressed in 293T cells and added to Ni-NTA
bound HIV-IN expressed in E. coli (Fig. 8A).
In support of the model, by a direct
Chapter II
35
measurement of the relative light units from
the immunoblot A3C was found to interact
10-fold stronger to the HIV-IN compared to
Vpr.3C (Fig. 8B).
Fig. 6: OptiPrep™ gradients of HIV particles show no difference in sub-viral localization of A3C and Vpr.3C.
(A) Determination of the RT activity in pg/mL (black lines) to display the localization the viral core containing
fractions. The density (in g/mL) of each gradient fractions is displayed as dotted lines. (B) Immunoblot analysis
to detect A3C and Vpr.3C (α-HA) in the different gradient fractions. An antibody against the VSV-glycoprotein
(α-VSV-G) was used to determine the fractions that contain free viral proteins.
DISCUSSION
The antiviral activity of the ubiquitously
expressed A3C (21-23)(18-20) is restricted to
SIV ∆vif and retroelements. The Vif-
independent resistance of HIV-1 to A3C
implicated an unknown viral mechanism to
counteract A3 proteins. Unfortunately this
study could not entirely elucidate the
presumed novel viral mechanism, but the
results indicate a new role for IN as an
antagonist against A3C.
A prerequisite for A3-mediated viral
restriction is the incorporation of A3 proteins
in newly budding virions. Although A3C is
incorporated into HIV ∆vif particles no
inhibition of viral infectivity could be
observed. Fusion of the protein of HIV-1Vpr to
A3C resulted in a restriction of HIV-1. The
mechanism by which Vpr.3C and 3C.Vpr
inhibit the viral infectivity is so far not entirely
understood. However, this restriction of HIV-1
by Vpr-A3C fusion proteins could not be
counteracted through co-expression of Vif. It
is thus possible that a sterical hindrance
through the fused Vpr protein masked the Vif
recognition site in A3C. But the Vif resistance
to Vpr.3C and 3C.Vpr is difficult to explain.
An inhibition of Vif binding by sterical
hindrance by both, the N-terminal and the C-
terminal Vpr domain is mechanistically not
obvious and a valuable topic for further
experiments. Vpr itself behaved inert in this
system and an effect of Vpr on the cell cycle
(11, 12) can be excluded, because the fusion
of Vpr to the enzymatic inactive A3C mutant
C98S had no influence on viral infectivity.
The Vpr fusion did also not change the sub-
viral localization of A3C in contrast to the
results observed for A3A (7).
Chapter II
36
Fig. 7: A3C interacts with integrase of HIV. (A)
Immunblot analysis of HIV and SIV integrase
expression in E. coli and their binding to Ni-NTA
agarose via a C-terminal His-tag using an α-His
antibody. (B) Binding of HA-tagged A3C to the
integrases was detected using an α-HA antibody.
To investigate the sub-viral localization of
A3C and Vpr.3C a fractionation via a 20-60%
OptiPrep™ gradient was performed.
Unexpectedly, both A3C and the Vpr.3C
fusion protein were localized in fractions with
similar density. Further a peak in RT activity
could be observed in these fractions, strongly
suggesting presence of viral core structures.
These results exclude a random packaging of
A3C into newly budding virions as a result of
intensive overexpression in producer cells,
because A3C and Vpr.3C are not found in
fractions of free viral proteins after detergent
treatment such as the envelope proteins.
Therefore the incorporation of A3C and
Vpr.3C into HIV virions must be a specific
and directed process. As differences in the
sub-viral localization cannot answer the
question why Vpr.3C is antiviral active and
A3C is not, one might speculate on Vpr.3C
dimers which have a higher affinity to each
other. This would be a reasonable
explanation, as the Vpr protein itself forms
dimers (24) and dimerization was shown to
be crucial for antiviral activity of A3C (see
Chapter I). Obviously HIV must have another
factor besides Vif that is counteracting the
antiviral activity of A3C. This so far unknown
factor could destroy A3C dimers but is not
potent enough to disrupt dimers of the
Vpr.3C fusion protein. This factor must be
Fig. 8: HIV-IN binds A3C more efficiently than
Vpr.3C. (A) Immunoblot analysis of A3C and
Vpr.3C expression in 293T cells (lysate) and the
binding of both proteins to HIV-IN (bound) using
an α-HA antibody. A3C and Vpr.3C were bound to
comparable amounts of HIV-IN as determined via
immunoblot analysis of Ni-NTA bound integrase
using an α-His antibody. (B) Relative binding of
A3C and Vpr.3C to the HIV-IN was quantified by
direct measurement of the relative light units from
the immunoblots in (A). Binding of A3C to HIV-IN
was set to 100%.
encoded by HIV, but not by SIV and this
factor must show differences in the
interaction with A3C and Vpr.3C. One of the
first candidates was the Viral protein U (Vpu),
as this protein is encoded by HIV, but not by
SIV. Unfortunately A3C was not able to
restrict the infectivity of vpu-deficient HIV
(data not shown). After screening different
other viral proteins for their physical
interaction with A3C, the most promising
candidate for an HIV encoded antagonist
against A3C was thought to be the HIV-1
Chapter II
37
integrase. Interaction studies of A3C with the
integrases of HIV and SIV showed that A3C
only interacts with the HIV integrase. This
emphasizes the idea of an antagonist,
because HIV ∆vif is not restricted by A3C, in
contrast to SIV ∆vif. Additionally, the amino
acid seqeuence of the HIV and SIV
integrases are 57.8% identical, which
facilitates different binding properties.
Furthermore the fact, that the HIV restricting
fusion protein Vpr.3C shows a 10-fold lower
interaction to the HIV-IN supports this
hypothesis. As interaction studies can only
provide an indication on a possible function,
an obvious experiment was to create an HIV
∆vif luc reporter construct that encodes for
the SIV-IN instead of its own integrase. This
chimeric virus should therefore be restricted
by A3C. Unfortunately, such a chimeric virus
is unable to integrate into the host cell
genome.
Future experiments are necessary to
emphazise the idea of the HIV integrase as
antagonist of A3C mediated viral restriction.
A fusion of Vpr to HIV-IN could deliver the
protein together with A3C into SIV ∆vif
particles, for instance. The presence of the
HIV-IN in those SIV particles should then
counteract the restriction of A3C against
SIV ∆vif. As inhibition of dimerization could
be an A3C neutralizing effect of the HIV-IN,
experiments targeting the dimerization of
A3C and Vpr.3C in presence and absence of
the integrases of HIV and SIV could be
performed.
Altogether, these indications and early
findings clearly lead to the integrase as
potential A3C antagonist of HIV-1.
Materials and Methods
Plasmids. HA-tagged A3C and A3G constructs
have been described previously (4, 25). To
express the fusion proteins of VprHIV-1 and A3C
the cDNA of Vpr of HIV-1 was fused either to the
5´ start (Vpr.3C) or to the 3´ end (3C.Vpr) of A3C
with a Gly4-Ser-linker in between. The fusion PCR
was performed using KOD XL Polymerase
(Novagen) according to the company’s
instructions. The final fusion product was cloned
into pcDNA 3.1 (+) from Invitrogen using BamHI
and NotI restriction sites. Primer sequences are
listed in Suppl. table S2. Viral vectors were
produced by cotransfecting pSIVagm Δvif luc (17)
or pNL 4-3 Δvif luc (26), pMD.G (a VSV.G
expression plasmid), and supplemented (or not)
with pcVif-SIVagm-V5 (27) or pc.Vif.HIV1-V5 (28).
Protein expression and virus production
293T cells were co-transfected with A3C, A3G or
Vpr-fusion protein expression plasmids and
pSIVagm Δvif luc or pNL 4-3 Δvif luc using
Lipofectamin LTX (Invitrogen). To pseudotype the
viral particles with the VSV-glycoprotein pMD.G
was added. Vif expression plasmids were co-
transfected to put back Vif to counteract A3
activity. 2days post-transfection virions and cells
were harvested. Cells were immediately lysed in
the appropiate volume of RIPA lysis buffer (25mM
Tris, pH 8, 137mM NaCl, 1% glycerol, 0.1% SDS,
0.5% sodiumdeoxycholate, 1% NP-40) for 10 min
on ice and lysates were clarified by centrifugation
in a table top centrifuge for 10 min at 4°C with
14.000rpm. Cell lysates were subsequently used
for immunoblot analysis to detect protein
expression using an α-HA antibody (1:10.000
dilution, Covance) and α-mouse horseradish
peroxidase (1:7.500, Amersham Biosciences).
Tubulin detection served as loading control
(1:10.000, Sigma). Binding was visualized by ECL
plus (Amersham Biosciences). Harvested viral
particles were filtered using 0.45µm MiniSart
Filters (Sartorius) and downstream used for
luciferase reportervirus assays or to analyze the
incorporation of A3C, A3G and the fusion proteins
into those particles.
Chapter II
38
Luciferase reporter virus assay. To measure the
infectivity of viral particles produced in the
presence and absence of co-transfected proteins,
the 2 days post transfection harvested virions
were normalized by RT-concentration and used to
transduce 2 × 103 HOS cells in a 96-well dish.
Three days after transduction, intracellular
luciferase activity was quantified using Steady Lite
HTS (Perkin Elmer). Data are presented as the
average counts per second of the triplicates ±
standard deviation. RT concentration was
quantified using the Lenti-RT Activity Assay
(Cavidi Tech).
Incorporation of A3 proteins into virions. To
detect co-transfected proteins in viral particles, the
2days post transfection harvested, filtered and RT
-normalized virions were precipitated by
ultracentrifugation over a 20% sucrose cushion
and lysed using RIPA lysis buffer. These lysates
of whole viral particles were then analyzed by
immunoblot analysis for incorporation of A3
proteins with an α-HA antibody as described
above. Equal loading amounts were analyzed
using the capsid monoclonal antibody AG 3.0 (29)
in a 1:250 dilution. This antibody detects HIV p24
and cross-reacts with SIV p27 and was therefore
used in both cases.
Determination of sub-viral localization.
Harvested and filtered virion containing
supernatants were pelleted over a 20% sucrose
cushion by ultracentrifugation. The particles were
then resuspended in 1 mL PBS. An 11 mL density
gradient consisting of 20-60% OptiPrep™ (Sigma)
was produced. On top of the gradient was a 1 mL
layer of 10% OptiPrep™ and 1% TritonX-100
followed by 2mL layer with 5% OptiPrep™ to build
a boarder to the top 1 mL layer of resuspended
viral particles. Ultracentrifugation of these
gradients was performed for 20 hrs at 4°C with
80.000 rcf. Subsequently eleven 1 mL fractions
from bottom to top of the gradient were collected
and either used for determination of RT
concentration or for immunoblot analysis. The
density of each fraction was calculated through
measuring the refraction index at 20°C. Specific
RT activity was quantified using the Lenti-RT
Activity Assay (Cavidi Tech). Localization of A3
proteins in the different fractions of the gradient
was determined by immunoblot using the
described α-HA antibody. To detect the envelope
proteins outside the viral core an anti-VSV-G
antibody (Sigma; 1:20.000) was used.
A3C-Integrase interaction. The integrases were
expressed from the plasmids pKB-IN6H (HIV-1 IN)
(30) and pCP-SIVIN6H (SIVmac-IN; a gift from P.
Cherepanov) in in the E. coli strain Rosetta BL21
DE3. Empty expression vector pET-20b(+) (31)
was used as negative control. Bacteria were
transformed with the respective plasmid DNA and
grown upon an OD of 0.5 at 600nm and
heterologous protein expression was induced with
0.1mM isopropyl 1-thio-β-D-galactopyranoside
(IPTG). After over night incubation at room
temperature cells were harvested, resuspended in
binding buffer (50mM NaxHyPO4, 300mM NaCl,
10mM imidazole) and lysed by sonification. Ni-
NTA Agarose (Invitrogen) was incubated with
clarified lysates for 2 hours at 4 °C and washed 5
times with washing buffer (50mM NaxHyPO4,
300mM NaCl, 20mM imidazole). 293T cells were
transfected with A3C or Vpr.3C expression
plasmids and cells were harvested 2 days post
transfection, resuspended in binding buffer and
lysed by sonification. The Ni-NTA agarose bound
integrases were incubated with clarified lysates
from 293T cells expressing A3C or Vpr.3C for
3 hours at 4°C and washed 5 times with washing
buffer. After washing Ni-NTA beads were
incubated with elution buffer (50mM NaxHyPO4,
300mM NaCl, 250mM imidazole) for 10min at
room temperature to elute bound proteins.
Proteins in the lysates and the elution fraction
were analyzed by immunoblotting using an α-His-
C-terminal antibody (Invitrogen; 1:4000)
(integrases) or the described α-HA antibody (A3C
and Vpr.3C). Light units of the elution fractions
were directly quantified from membranes
incubated with ECL plus using the Lumianalyst 3.0
software (Roche).
Chapter II
39
References
1. Holmes RK, Malim MH & Bishop KN (2007) APOBEC-mediated viral restriction: not simply editing? Trends Biochem. Sci. 32, 118-128.
2. Sheehy AM, Gaddis NC, Choi JD & Malim MH (2002) Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral Vif protein. Nature 418, 646-650.
3. Zheng YH et al. (2004) Human APOBEC3F is another host factor that blocks human immunodeficiency virus type 1 replication. J. Virol. 78, 6073-6076.
4. Muckenfuss H et al. (2006) APOBEC3 proteins inhibit human LINE-1 retrotransposition. J. Biol. Chem. 281, 22161-22172.
5. Yu Q et al. (2004) APOBEC3B and APOBEC3C are potent inhibitors of simian immunodeficiency virus replication. J. Biol. Chem. 279, 53379-53386.
6. Bourara K, Liegler TJ & Grant RM (2007) Target cell APOBEC3C can induce limited G-to-A mutation in HIV-1. PLoS. Pathog. 3, 1477-1485.
7. Aguiar RS, Lovsin N, Tanuri A & Peterlin BM (2008) Vpr.A3A chimera inhibits HIV replication. J. Biol. Chem. 283, 2518-2525.
8. Cohen EA et al. (1990) Identification of HIV-1 vpr product and function. J. Acquir. Immune. Defic. Syndr. 3, 11-18.
9. Heinzinger NK et al. (1994) The Vpr protein of human immunodeficiency virus type 1 influences nuclear localization of viral nucleic acids in nondividing host cells. Proc. Natl. Acad. Sci. U. S. A 91, 7311-7315.
10. Subbramanian RA et al. (1998) Human immunodeficiency virus type 1 Vpr localization: nuclear transport of a viral protein modulated by a putative amphipathic helical structure and its relevance to biological activity. J. Mol. Biol. 278, 13-30.
11. Jowett JB et al. (1995) The human immunodeficiency virus type 1 vpr gene arrests infected T cells in the G2 + M phase of the cell cycle. J. Virol. 69, 6304-6313.
12. Planelles V et al. (1996) Vpr-induced cell cycle arrest is conserved among primate lentiviruses. J. Virol. 70, 2516-2524.
13. Kondo E, Mammano F, Cohen EA & Gottlinger HG (1995) The p6gag domain of human immunodeficiency virus type 1 is sufficient for the incorporation of Vpr into heterologous viral particles. J. Virol. 69, 2759-2764.
14. Paxton W, Connor RI & Landau NR (1993) Incorporation of Vpr into human immunodeficiency virus type 1 virions: requirement for the p6 region of gag and mutational analysis. J. Virol. 67, 7229-7237.
15. Fletcher TM, III et al. (1997) Complementation of integrase function in HIV-1 virions. EMBO J. 16, 5123-5138.
16. Wu X et al. (1997) Functional RT and IN incorporated into HIV-1 particles independently of the Gag/Pol precursor protein. EMBO J. 16, 5113-5122.
17. Mariani R et al. (2003) Species-specific exclusion of APOBEC3G from HIV-1 virions by Vif. Cell 114, 21-31.
18. Dettenhofer M & Yu XF (1999) Highly purified human immunodeficiency virus type 1 reveals a virtual absence of Vif in virions. J. Virol. 73, 1460-1467.
19. Mouland AJ et al. (2000) The double-stranded RNA-binding protein Staufen is incorporated in human immunodeficiency virus type 1: evidence for a role in genomic RNA encapsidation. J. Virol. 74, 5441-5451.
20. Wang X, Dolan PT, Dang Y & Zheng YH (2007) Biochemical differentiation of APOBEC3F and APOBEC3G proteins associated with HIV-1 life cycle. J. Biol. Chem. 282, 1585-1594.
21. Chiu YL & Greene WC (2008) The APOBEC3 cytidine deaminases: an innate defensive network opposing exogenous retroviruses and endogenous retroelements. Annu. Rev. Immunol. 26, 317-353.
22. Koning FA et al. (2009) Defining APOBEC3 expression patterns in human tissues and hematopoietic cell subsets. J. Virol. 83, 9474-9485.
Chapter II
40
23. Refsland EW et al. (2010) Quantitative profiling of the full APOBEC3 mRNA repertoire in lymphocytes and tissues: implications for HIV-1 restriction. Nucleic Acids Res. 38, 4274-4284.
24. Zhao LJ, Wang L, Mukherjee S & Narayan O (1994) Biochemical mechanism of HIV-1 Vpr function. Oligomerization mediated by the N-terminal domain. J. Biol. Chem. 269, 32131-32137.
25. Lochelt M et al. (2005) The antiretroviral activity of APOBEC3 is inhibited by the foamy virus accessory Bet protein. Proc. Natl. Acad. Sci. U. S. A 102, 7982-7987.
26. Loewen N et al. (2003) FIV Vectors. Methods Mol. Biol. 229, 251-271.
27. Perkovic M et al. (2008) Species-specific inhibition of APOBEC3C by the prototype foamy virus protein Bet. J. Biol. Chem.
28. Zielonka J et al. (2010) Vif of feline immunodeficiency virus from domestic cats protects against APOBEC3 restriction factors from many felids. J. Virol. 84, 7312-7324.
29. Simm M et al. (1995) Aberrant Gag protein composition of a human immunodeficiency virus type 1 vif mutant produced in primary lymphocytes. J. Virol. 69, 4582-4586.
30. Maertens G et al. (2003) LEDGF/p75 is essential for nuclear and chromosomal targeting of HIV-1 integrase in human cells. J. Biol. Chem. 278, 33528-33539.
31. Cherepanov P (2007) LEDGF/p75 interacts with divergent lentiviral integrases and modulates their enzymatic activity in vitro. Nucleic Acids Res. 35, 113-124.
Chapter III
41
CHAPTER III
APOBEC3C mediated restriction of SIV is editing-independent
Henning Hofmann and Carsten Münk
Heinrich-Heine-University Düsseldorf, Clinic for Gastroenterology, Hepatology and Infectiology, Building
23.12_U1 Room 81, Moorenstr. 5, 40225 Düsseldorf, Germany
Data not published.
ABSTRACT
Different retroviruses and retroelements are
restricted by cellular cytidine deaminases of
the APOBEC3 protein family. These proteins
inhibit viral replication by deamination of
cytidines to uridines on ssDNA during
retroviral reverse transcription, a well
characterized mechanism called editing.
However, APOBEC3C’s (A3C) antiviral
activity against Simian immunodeficiency
virus (SIV) lacking its vif gene (SIV Δvif) was
not caused by editing.
This study aimed to find an alternative
restriction mechanism of A3C against
SIV Δvif. Different steps in viral replication
that are potential targets of APOBEC3
mediated restriction were examined. The
results demonstrate, that A3C neither
induces RNA-editing nor inhibits reverse
transcription. Thus, A3C likely uses an
additional, so far unidentified mechanism to
restrict SIV Δvif.
INTRODUCTION
APOBEC3 proteins differ from each other in
having one or two Zn2+-coordinating (Z)
domains, that belong either to the Z1, Z2, or
Z3 type (1). The best described member of
the APOBEC3 protein family, APOBEC3G
(A3G), comprises an N-terminal Z2- and a C-
terminal Z1-domain. Only the C-terminal
domain mediates the enzymatic activity
whereby retroviruses and transposable
elements are inhibited (2-4). The restriction
pattern of A3G is much broader than that of
A3C (reviewed in (5)) which owns a single
Z2-domain. A3G inhibits HIV Δvif mainly by
deamination of cytidine to uridine of ssDNA
formed during reverse transcription (6-9).
These mutations either lead to degradation of
viral DNA by DNA base repair enzymes
(uracil DNA glycosylase and apurinic-
apyrimidinic endonuclease) (10), or to
extensive G to A hypermutations on the
coding strand of viral DNA. These editing
events could result in alterations of the amino
acid code or in introduction of inappropriate
translation termination codons (7-9, 11).
However, other mechanisms including
inhibition of reverse transcription or
Chapter III
42
integration have been discussed recently.
(12-16).
This study aimed to answer two questions: a)
is the Zn2+-coordinating domain of A3C
responsible for its antiviral activity against
SIV ∆vif and b) what is the restriction
mechanism of A3C.
RESULTS
The Zn2+-coordinating motif is the active
site of A3C
The two Zn2+-coordinating domains of
APOBEC3G (A3G) display different
functional properties. The N-terminal Z2-
domain (NTD) is responsible for virion
incorporation of the protein and the C-
terminal Z1-domain (CTD) was shown to be
the enzymatic active site of A3G (2, 4). To
determine the function of the single Z2-
domain of A3C the zinc ion-binding amino
acids were mutated as shown in Fig. 1.
Fig. 1: Schematic overview of the point mutations
in the Zn2+-coordinating domain (red) of A3C
(grey). The single amino acids changes are shown
as red letters.
These A3C mutants were tested against SIV
∆vif using luciferase reporter virus. VSV-G
pseudotyped SIVagm luc ∆vif virions were
produced in 293T cells in presence of wt or
mutant A3C and in presence or absence of
the Vif protein of SIV.
Fig. 2: Mutations in the Zn2+ - coordinating domain results in loss of function proteins. (A) Antiviral activity of
A3C wt and mutant proteins against SIVagm Δvif luc and SIVagm luc, compared to non-transduced cells (mock)
and vector only (control without A3C). (B) Immunoblot analysis of the expression, Vif-dependent degradation
in producer cells and incorporation into virions of A3C wt and mutant proteins. Tubulin and p27 (capsid)
served as loading control.
Chapter III
43
Normalized viral particles were used to
transduce human osteosarcoma (HOS) cells.
Intracellular luciferase activity in those cells
determined the infectivity of used virions.
Only wt A3C could reduce viral infectivity
displayed as a decrease in luciferase activity
(Fig. 2A). Mutation of any Zn2+-coordinating
amino acid resulted in loss-of-function
proteins, likely by destroying the active site.
As these mutants were recognized by the Vif
protein of SIV inducing their proteasomal
degradation (Fig. 2B) they should not be
subject to severe misfolding. Additionally,
virion incorporation of A3C wt and mutant
proteins was examined as this is a
prerequisite for antiviral activity (Fig. 2B). The
mutants H66R, E68Q, C98S were
incorporated into viral particles similar to
wt A3C, indicating that the non-restricting
phenotype is not based on the exclusion of
the mutants from the virions. Although the
mutated proteins were packaged, mutations
impaired their enzymatic active domain.
Mutant C101S was packaged to a
significantly lower extent and for this protein
a low encapsidation into virions and/or the
damaged active site can explain its inactivity
against SIV Δvif.
Viral DNA is not deaminated by A3C As deamination of viral cDNA is the most
prominent mechanism of A3-mediated
retroviral restriction, we examined whether
DNA editing is also the reason for SIV ∆vif
inhibition by A3C. Therefore pseudotyped
SIV∆ vif luc virions, generated in presence or
absence of A3C, were used to transduce
HOS cells. At indicated time points post
infection total DNA was isolated from those
cells and a 600bp long fragment of the SIV
gag gene was amplified via PCR. The
resulting products were then sequenced and
analyzed for G to A mutations (Fig. 3). The
results display that A3C does not deaminate
the viral cDNA of SIV ∆vif. DNA isolated at 4,
9, 12 or 24 hours post infection in the
presence of A3C showed G to A mutation
ratios similar to that of the negative control
(SIV ∆vif only) (Fig. 3). As a positive control,
HOS cells were transduced with virions
produced in the presence of A3G. PCR
products of this infection showed the
expected extensive hypermutation of the viral
genome caused by A3G.
A3C does not edit viral RNA As deamination of viral cDNA was likely not
involved in the antiviral activity of A3C,
another possible mechanism was editing of
viral RNA prior to reverse transcription. Total
RNA was isolated from cells transduced with
VSV-G pseudotyped SIV∆vif luc virions
produced in the presence or absence of A3C.
After cDNA synthesis from total RNA a region
including the primer binding site (PBS) was
amplified and analyzed for mutations.
Primers amplifying the PBS were chosen, as
the initial step of reverse transcription starts
with binding of the tRNA to the PBS (Fig. 4A)
(17). The sequences (n=7) of SIV made in
the presence of A3C showed that more non-
specific (non C to U) mutations were
detectable than mutations that could have
resulted from A3C’s enzymatic activity (Fig.
4B). Analyzing sequences after infection with
SIV made in the absence of A3C ended up
with a similar result. Taken together, A3C
does neither deaminate viral cDNA nor viral
RNA.
Chapter III
44
Fig. 3: A3C does not deaminate the genome of SIV ∆vif. Each G to A mutation is depicted as a dot. Five
different clones were analyzed after different time points post infection. Editing ratios were calculated as the
percentage of G to A mutations per 100 nt (grey). A3G served as positive control, SIV ∆vif only displays the
background of A3-independend mutations.
A3C does not block reverse transcription
The products of retroviral reverse
transcription (RT) can be divided in early and
late RT products according to their
appearance during the retroviral life cycle
(18).
Fig. 4: No deamination of SIV RNA detectable in
the presence of A3C. (A) Primers are shown as
black arrows flanking the PBS in the SIV genome.
(B) Mutation ratios (in %) in presence (+) and
absence (w/o) of A3C were calculated as C to U
mutation on viral RNA per 100 nt.
To investigate, if the RT reaction is
influenced by A3C at all, we first analyzed
late RT products. HOS cells were transduced
with normalized amounts of virions (SIV∆vif
luc) produced with or without A3C and 2, 4,
24, 48 and 96 hours post infection total DNA
was isolated from these cells. Via quantitative
real time PCR the amount of viral, reverse
transcribed cDNA was quantified as fg per
100ng of total DNA using primers specific for
late RT products. The results shown in Fig. 5
indicate a delay in the appearance of late RT
products of SIVΔvif made in the presence of
A3C, whereas the overall amount of RT
products remained the same. To proof that
the observed A3C dependent decrease of
luciferase activity in SIV∆vif luc infected cells
after 3 days (Fig. 2) does not result from
delayed viral replication, intracellular
luciferase activity was also measured at day
Chapter III
45
4 post infection (Fig. 6). However, luciferase
activity of virions produced in presence of
A3C analyzed on day 3 and day 4 post
infection were comparable. Thus, the 10 –
100-fold antiviral effect of A3C likely cannot
be explained by inhibition of reverse
transcription.
DISCUSSION
The mechanisms responsible for the
retroviral restriction by APOBEC3 proteins
are still not completely solved. Besides
deamination of the viral genome, other
possibilities of A3-mediated restriction have
been discussed recently (12-16), such as
inhibition of reverse transcription or
integration. In two domain APOBEC3
proteins only one domain exhibits enzymatic
activity, whereas the other domain is
necessary for virion incorporation (2-4). Since
A3C comprises only a single Zn2+-
coordinating domain, it was unclear how the
protein manages both, encapsidation and
antiviral activity.
Fig. 5: Late reverse transcription products appear slightly later in the presence of A3C. Quantitative real-time
PCR analysis of SIVagm Δvif luc late RT products after indicated time points post infection in target cells.
Amount of reverse transcribed products was quantified as fg per 100ng of total isolated DNA.
To answer whether the Zn2+-coordinating
domain is the enzymatic active site of A3C,
the crucial amino acids within this domain
were mutated. These mutations resulted in a
complete loss of antiviral activity although
most of the mutants were efficiently
incorporated, proofing that the single Zn2+-
coordinating domain is responsible for
enzymatic activity. Later it was shown that a
binding pocket proximal to this active site is
crucial for RNA-dependent incorporation of
A3C into viral particles (see Chapter I).
Although in a bacterial deamination assay it
was shown that A3C has the intrinsic
capacity to deaminate cytidines (see Chapter
I), there was no editing detectable in the
context of SIV ∆vif infections in human cells.
This result is in contrast to published data
(19), where a very low editing ratio of A3C
has been described. One could speculate
about A3C inducing mutations resulting in
viral restriction that might have occurred
elsewhere in the viral genome instead of the
analyzed part of the gag gene. This
Chapter III
46
possibility is very unlikely, as A3 proteins
have certain recognition sequences (20) that
consist of 2-4 nt being found within the whole
viral genome. Also A3G clearly showed a
strong deamination pattern within this part of
the SIV genome. As the experimental results
argue against cytidine deamination being the
main mechanism of A3C-mediated restriction
of SIV ∆vif, other possible molecular
consequences of A3C activity were analyzed.
To investigate inhibition of reverse
transcription, two strategies were pursued.
The first was RNA editing at the PBS. The
RT reaction is initiated through binding of a
tRNA at the PBS. The tRNA binding could be
inhibited by destroying the recognition
sequence of the PBS and reverse
transcription would be blocked at the very
first step. In agreement that A3 proteins act
on ssDNA (20), RNA editing of A3C could not
Fig. 6: Measurement of the antiviral activity of A3C against Δvi SIVagm luc at 3 or 4 days post infection (p. i.) in
presence (white bars) or absence (black bars) of the Vif protein SIVagm showed no significant differences.
be proven. The second approach was to
examine whether the RT reaction was
influenced through the presence of A3C. For
A3G a decrease in HIV-1 reverse
transcription products was determined when
the protein was present (14, 21). It remains
unclear whether this inhibition is caused by
direct physical interaction with the viral
enzyme reverse transcriptase. A3G could
also bind to viral cDNA after first strand
synthesis and thereby sterically block the
synthesis of the second cDNA strand after
strand transfer (15). To investigate whether
A3C inhibits the RT reaction during infection,
a quantitative PCR approach for late RT
products was chosen amplifying a 600nt long
part of the gag gene that is reverse
transcribed at late stages of the dsDNA
synthesis. Quantifying these late occurring
RT products would show whether reverse
transcription is inhibited at all. The results
displayed a delay in the appearance of those
late RT products caused by A3C. But as the
overall amount of viral cDNA produced in
presence or absence of A3C was
comparable, it can be concluded that A3C
does not inhibit reverse transcription. In the
presence of A3C the RT products are
moderately delayed. But this delay in the RT
reaction did not correlate with a “restored”
infectivity at later time points post infections.
More experiments are required to test,
Chapter III
47
whether the next steps in replication, nuclear
import and integration are affected by A3C.
Materials and Methods
Plasmids. HA-tagged A3C and A3G constructs
have been described previously (22, 23). The
point mutations of A3C were inserted using site
directed mutagenesis. Therefore A3C expression
plasmid was used as template for a PCR reaction
amplifying the whole plasmid with primers carrying
those nucleotides to be changed. Primer
sequences: H66R-5’: 5’-tctgcacgacaatgggtctca-3’,
H66R-3’: 5’-gagacccgttgtcatgcaga-3’, E68Q-5’: 5’- cacctttgtgcatgacaatg-3’, E68Q-3’: 5’-atgcacaaag
gtgcttcctc-3’, C98S-5’: 5’-tctggagaagggctccaagat-
3’, C98S-3’: 5’-gcccttctccagactgtgca-3’, C101S-5’:
5’-ctgcagagtctgggcaaggg-3’ and C101S-3’: 5’- cagactctgcaggggaggtg-3’. After PCR the products
were digested with DpnI to disrupt methylated,
parental DNA and transformed into E. coli strain
Top 10 (Invitrogen). Mutations were verified by
sequencing. Viral vectors were produced by co-
transfection of pSIVagm Δvif luc (24) or pNL 4-3
Δvif luc (25), pMD.G, a VSV-G expression
plasmid, and supplemented (or not) with pcVif-
SIVagm-V5 (26) or pc.vif.HIV1-V5 (27).
Protein expression and virus production. 293T
cells were co-transfected with A3C expression
plasmids and pSIVagmΔvif luc or pNL 4-3Δvif luc
using Lipofectamine LTX (Invitrogen). To
pseudotype the viral particles with the VSV-
glycoprotein pMD.G was co-transfected, too. Vif
expression plasmids were co-transfected to
counteract A3 activity. 2 days post transfection
virions and cells were harvested. Cells were
immediately lysed in an appropriate volume of
RIPA lysis buffer (25mM Tris, pH 8, 137mM NaCl,
1% glycerol, 0.1% SDS, 0.5%
sodiumdeoxycholate, 1% NP-40) for 10min on ice
and lysates were clarified by centrifugation in a
table top centrifuge for 10min at 4°C with
14.000rpm. Cell lysates were then used for
immunoblot analysis to detect protein expression
using an anti (α)-HA antibody (1:10.000, Covance)
and α-mouse horseradish peroxidase (1:7.500,
Amersham Biosciences). α -Tubulin was used as
loading control and detected using α-tubulin
antibody (1:10.000, Sigma). Signals were
visualized by ECL plus (Amersham Biosciences).
Harvested viral particles were filtered using
0.45µm MiniSart Filters (Sartorius) and
downstream used for luciferase reporter virus
infections or to analyze the incorporation of A3C
wt or mutant proteins into those particles.
Luciferase reporter virus assay. To measure the
infectivity of viral particles produced in the
presence and absence of co-transfected proteins,
2 days post transfection virions were harvested,
normalized by RT-concentration and used to
transduce 2×103 HOS cells in a 96-well dish.
Three days after infection, intracellular luciferase
activity was quantified using Steady Lite HTS
(Perkin Elmer). Data are presented as the average
counts per second of the triplicates ± standard
deviation. RT concentration was quantified using
the Lenti-RT Activity Assay (Cavidi Tech).
Incorporation of A3 proteins into virions.
To detect co-transfected proteins in viral particles,
the 2 days post transfection harvested, filtered and
RT -normalized virions were precipitated by
ultracentrifugation over a 20% sucrose cushion
and lysed using RIPA lysis buffer. These lysates
of whole viral particles were then analyzed by
immunoblot analysis for incorporation of A3
proteins with an α-HA antibody as described
above. Equal loading amounts were analyzed
using the capsid monoclonal antibody AG 3.0 (28)
(1:250). This antibody detects HIV p24 and cross-
reacts with SIV p27 and was therefore used in
both cases.
DNA Deamination assay. To detect deamination
of viral cDNA, 2 days post transfection virions
were harvested, filtered and RT normalized, as
well as incubated with DNaseI (Roche) to remove
residual plasmid DNA. HOS cells were transduced
with these virions and 4, 9, 12 and 24 hours post
infection total DNA was isolated using DNeasy
Blood & Tissue Kit (QIAGEN). 300ng of total DNA
was used as template to amplify a part of the gag
gene with the primers CM101 (5‘-
caggctgagaaatctccagcag-3‘) and CM102 (5‘-
ccatgtctgccactaggtgtcgc-3‘) using Taq polymerase
Chapter III
48
(Fermentas). The PCR products were cloned into
pJET1.2 cloning vector (Fermentas) and
sequenced. G to A mutations were detected by
aligning these sequences with the SIVagmTAN-1
gag sequence. Editing ratios were calculated as G
to A mutations per 100nt analyzed.
RNA Deamination assay. Infection of HOS cells
was performed similar to the DNA deamination
assay. 6 hours post infection total RNA was
isolated using RNeasy Kit (QIAGEN) and reverse
transcribed into total cDNA with Super Script III
Reverse Transcriptase (Invitrogen). 500ng of
cDNA were used for the PCR reaction with
primers PBS5´ (5‘-cttaagagtctatctgagcaag-3‘) and
PBS3´ (5‘-gtaattaccatgtctgccact-3‘) flanking the
primer binding site. PCR products were analyzed
similar to the DNA deamination assay.
Quantitative PCR to detect late RT products.
HOS cells were transduced with SIV Δvif virions
produced in presence and absence of A3C. At 2,
4, 9, 24, 48 and 96 hours post infection total DNA
was isolated using DNeasy Blood & Tissue Kit
(QIAGEN). 100ng of each total DNA were used for
a quantitative real-time PCR to detect late RT
products with the primers: CS: 5’-cactcgg
cactgtcaggga-3’ and CR: 5’-ggttctagcgggctcaata
cttctat-3’. Specificity of the products was
determined by use of Fluorescein (Fl) and Light
Cycler Red 640 (LC) labeled hybridization probes:
C-Fl: 5’-tgatactttttctttccgttcgggcg—Fl-3’ and 5’-
LC—aagcgtattttctcaaatgtgtccaaattcc-3’. The
following cycling conditions were used: 50 cycles
with 95°C for 10sec, 57°C for 20sec and 72°C for
10sec in Light Cycler 3 instrument from Roche.
The amount of late RT products in fg per 100ng
was determined by standardization of each run
with a serial dilution of pSIVagm Δvif plasmid.
Therefore 10fg, 100fg, 1pg and 10pg of plasmid
DNA were used.
References
1. LaRue RS et al. (2008) The artiodactyl
APOBEC3 innate immune repertoire
shows evidence for a multi-functional
domain organization that existed in the
ancestor of placental mammals. BMC.
Mol. Biol. 9, 104.
2. Gooch BD & Cullen BR (2008) Functional
domain organization of human
APOBEC3G. Virology 379, 118-124.
3. Hache G, Liddament MT & Harris RS
(2005) The retroviral hypermutation
specificity of APOBEC3F and
APOBEC3G is governed by the C-
terminal DNA cytosine deaminase
domain. J. Biol. Chem. 280, 10920-
10924.
4. Navarro F et al. (2005) Complementary
function of the two catalytic domains of
APOBEC3G. Virology 333, 374-386.
5. Holmes RK, Malim MH & Bishop KN
(2007) APOBEC-mediated viral
restriction: not simply editing? Trends
Biochem. Sci. 32, 118-128.
6. Bishop KN et al. (2004) Cytidine
deamination of retroviral DNA by diverse
APOBEC proteins. Curr. Biol. 14, 1392-
1396.
7. Lecossier D, Bouchonnet F, Clavel F &
Hance AJ (2003) Hypermutation of HIV-1
DNA in the absence of the Vif protein.
Science 300, 1112.
8. Mangeat B et al. (2003) Broad
antiretroviral defence by human
APOBEC3G through lethal editing of
nascent reverse transcripts. Nature 424,
99-103.
Chapter III
49
9. Zhang H et al. (2003) The cytidine deaminase
CEM15 induces hypermutation in newly
synthesized HIV-1 DNA. Nature 424, 94-
98.
10. Yang B et al. (2007) Virion-associated
uracil DNA glycosylase-2 and
apurinic/apyrimidinic endonuclease are
involved in the degradation of
APOBEC3G-edited nascent HIV-1 DNA.
J. Biol. Chem. 282, 11667-11675.
11. Harris RS et al. (2003) DNA deamination
mediates innate immunity to retroviral
infection. Cell 113, 803-809.
12. Bishop KN, Holmes RK & Malim MH
(2006) Antiviral potency of APOBEC
proteins does not correlate with cytidine
deamination. J. Virol. 80, 8450-8458.
13. Guo F et al. (2006) Inhibition of formula-
primed reverse transcription by human
APOBEC3G during human
immunodeficiency virus type 1
replication. J. Virol. 80, 11710-11722.
14. Holmes RK, Koning FA, Bishop KN &
Malim MH (2007) APOBEC3F can inhibit
the accumulation of HIV-1 reverse
transcription products in the absence of
hypermutation. Comparisons with
APOBEC3G. J. Biol. Chem. 282, 2587-
2595.
15. Mbisa JL et al. (2007) Human
immunodeficiency virus type 1 cDNAs
produced in the presence of APOBEC3G
exhibit defects in plus-strand DNA
transfer and integration. J. Virol. 81,
7099-7110.
16. Mbisa JL, Bu W & Pathak VK (2010)
APOBEC3F and APOBEC3G inhibit HIV-
1 DNA integration by different
mechanisms. J. Virol. 84, 5250-5259.
17. Harrich D & Hooker B (2002) Mechanistic
aspects of HIV-1 reverse transcription
initiation. Rev. Med. Virol. 12, 31-45.
18. Freed EO (2001) HIV-1 replication.
Somat. Cell Mol. Genet. 26, 13-33.
19. Yu Q et al. (2004) APOBEC3B and
APOBEC3C are potent inhibitors of
simian immunodeficiency virus
replication. J. Biol. Chem. 279, 53379-
53386.
20. Yu Q et al. (2004) Single-strand
specificity of APOBEC3G accounts for
minus-strand deamination of the HIV
genome. Nat. Struct. Mol. Biol. 11, 435-
442.
21. Bishop KN et al. (2008) APOBEC3G
inhibits elongation of HIV-1 reverse
transcripts. PLoS. Pathog. 4, e1000231.
22. Lochelt M et al. (2005) The antiretroviral
activity of APOBEC3 is inhibited by the
foamy virus accessory Bet protein. Proc.
Natl. Acad. Sci. U. S. A 102, 7982-7987.
23. Muckenfuss H et al. (2006) APOBEC3
proteins inhibit human LINE-1
retrotransposition. J. Biol. Chem. 281,
22161-22172.
24. Mariani R et al. (2003) Species-specific
exclusion of APOBEC3G from HIV-1
virions by Vif. Cell 114, 21-31.
25. Loewen N et al. (2003) FIV Vectors.
Methods Mol. Biol. 229, 251-271.
26. Perkovic M et al. (2008) Species-specific
inhibition of APOBEC3C by the prototype
foamy virus protein Bet. J. Biol. Chem.
Chapter III
50
27. Zielonka J et al. (2010) Vif of feline
immunodeficiency virus from domestic
cats protects against APOBEC3
restriction factors from many felids. J.
Virol. 84, 7312-7324.
28. Simm M et al. (1995) Aberrant Gag
protein composition of a human
immunodeficiency virus type 1 vif mutant
produced in primary lymphocytes. J.
Virol. 69, 4582-4586.
Appendix
51
Supplementary Material -
CHAPTER I
Model Structure of APOBEC3C Reveals a Binding Pocket Modulating RNA Interaction Required for Encapsidation
Benjamin Stauch, Henning Hofmann, Mario Perković, Martin Weisel, Ferdinand Kopietz,
Klaus Cichutek, Carsten Münk, Gisbert Schneider Supplementary Table S1. PCR primer sequences.
Appendix
52
Additional Material and Methods
Model refinement and evaluation.
MODELLER 9v4 implements the dynamic programming algorithm by a variable gap penalty (1):
Based on the PDB template structures, gaps were penalized differentially and favored in solvent-
exposed loops and surface regions compared to secondary structure elements and core regions
and therefore placed in a preferential structural context. Penalties were given as a vector with
values previously suggested as optimal for the sequence identity found in the alignment (1).
REDUCE (2) was applied to add hydrogens to all initial models obtained by MODELLER 9v4 (24,
42). The models were subsequently energy-minimized, using the CHARMM22 forcefield (3),
generalized Born implicit solvent model and a gradient descent algorithm as implemented in MOE
2006.08 (Chemical Computing Group, Montreal, Canada): first hydrogens only, then sidechains,
finally backbone atoms, tethered by a force constant of 0.1. PROCHECK v.3.5 (4) was used to
evaluate both the minimized models and the template structures. The models with fewest violations
of backbone angles were equilibrated by Molecular Dynamics (MD) simulation. The MD simulations
were carried out by NAMD Scalable Molecular Dynamics (5) in a water sphere with harmonic
boundary conditions, CHARMM22 force field, and Langevin Dynamics in a canonic NVT ensemble,
2 fs timestep. After initial minimization for 10,000 discrete steps, the simulation was run for 20 ns at
310 K. The resulting trajectory was visualized and evaluated using VMD Visual Molecular
Dynamics (6). The ANOLEA software (27), implementing a knowledge-based residue-wise pseudo-
potential, was utilized to generate ”energy” profiles of the final model structures of A3C and the
respective PDB templates, using five-residue window averaging. Differences between superposed
structures were quantified as root mean square deviation (RMSD) of Cα atoms. Pictures of protein
models were generated using UCSF Chimera v1 (7).
Additional references 1. Madhusudhan MS, Marti-Renom MA, Sanchez R & Sali A (2006) Variable gap penalty for
protein sequence-structure alignment. Protein Eng Des Sel 19, 129-133.
2. Word JM, Lovell SC, Richardson JS & Richardson DC (1999) Asparagine and glutamine: using
hydrogen atom contacts in the choice of side-chain amide orientation. J. Mol. Biol. 285,
1735-1747.
3. McKerell AD et al. (1998) All-atom empirical potential for molecular modeling and dynamics
studies of proteins. J. Phys. Chem. B. 102, 3586-3616.
4. Laskowski RA, MacArthur MW, Moss DS & Thornton JM (1993) PROCHECK: a program to
check the stereochemical quality of protein structures. J. Appl. Cryst. 26, 283-291.
5. Phillips JC et al. (2005) Scalable molecular dynamics with NAMD. J. Comput. Chem. 26, 1781-
1802.
6. Humphrey W, Dalke A & Schulten K (1996) VMD: visual molecular dynamics. J. Mol. Graph. 14,
33-38.
7. Pettersen EF et al. (2004) UCSF Chimera-a visualization system for exploratory research and
analysis. J. Comput. Chem. 25, 1605-1612.
Appendix
53
Figure S1. Ramachandran plots of (a) human APOBEC2 (A2), chain B (crystal structure), (b) A2-
derived model structure of human APOBEC3C (A3C), (c) human APOBEC3G, C terminal domain
(A3G-CD) and (d) A3G-CD-derived model structure of A3C. Stereochemical regions are colored
red (most favourable), bright yellow (additionally allowed), pale yellow (generously allowed) and
white (disallowed). Amino acids violating stereochemistry are indicated as red squares and
numbered according to original PDB files / A3C amino acid sequence. Glycins are indicated as
black triangles, all remaining amino acids as black squares.
(A2 crystal structure: 87.6% most favored, 11.8% additionally allowed, 0.6% generously allowed,
0% disallowed; A3G-CD solution structure: 84.8, 11.0, 2.4, 1.8%, respectively)
(A2-derived model: 83.7% most favorable, 14.0% additionally allowed, 0.6% generously allowed,
1.8% disallowed; A3G-CD-derived model: 84.9, 9.9, 2.3, 2.9%, respectively, with all generously
and disallowed residues being in loop regions or the N-terminus modeled without template)
The global sequence identity of A3C and A2 is 30.2%, for secondary structure regions it is only
marginally higher (32.8%). The global sequence identity of A3C and A3G-CD is 40.5%, but
significantly higher (53.5%) for secondary structure regions.
Appendix
54
Figure S2. Alignment of human APOBEC3C (A3C) to human APOBEC2 (A2) (a), human
APOBEC3G, C-terminal domain (A3G-CD) (b) and human APOBEC3G, N-terminal domain (A3G-
ND) (c). Secondary structure of A2 and A3G-CD as assigned by DSSP (H, helix; E, strand; else
loop) are indicated above the sequences, if available. Identical amino acids are given in bold and
denoted by ‘ | ‘, gaps in the respective sequence by ‘-’. The Zn2+-coordinating deaminase motif is
highlighted (red box). A2 and A3C share 30.2% identical amino acids, A3G-CD and A3C 40.5%,
A3G-ND and A3C 44.0%.
Appendix
55
Figure S3. ANOLEA-profiles for (a) the A2 crystal structure, (b) the A3G-CTD crystal structure, (c)
the A2-based model of A3C and (d) the A3G-CTD-based model structure of A3C. ANOLEA-
“energies” represent a knowledge-based pseudo-potential and are plotted for each sequence
position of the respective protein model.
Figure S4. Equilibration of human APOBEC3C (A3C). Development of the total (a), van-der-Waals
(b) and electrostatic energy (c) of the human A3C model over time. Energies are shown over the
first 25 picoseconds (ps). During the course of the total 20 ns of simulation, energies are stationary
(not shown).
Appendix
56
Figure S5. Evaluation of MD simulation of APOBEC structures. Development of the deviation
of the conformation of the crystal structure of A2 (a), the A2-derived model of A3C (b), the crystal
structure of A3G-CTD (c), and the A3G-CTD-derived model of A3C (d) with respect to the starting
conformation over the course of the simulation, measured in Å root mean square deviation
(RMSD). Simulations have been replicated three times (black, blue, red curve). Experimental (blue)
and simulated B-factors (red) of C-alpha atoms of human APOBEC2, chain B (e) and human
APOBEC3G, C-terminal domain (f).
e
f
a
b
c
d
Appendix
57
Figure S6. (a) Quantification of light units from immunoblot (IB) of bound V5-tagged wt A3C to
indicated co-transfected HA-tagged wt A3C or W74A. Numbers indicate the signal intensity of V5-
tagged wt A3C bound to W74A relative to the amount of V5-tagged wt A3C bound to HA-tagged wt
A3C. Background interaction of V5-tagged wt A3C with anti-HA affinity beads was determined
equally. (b) Immunoblot analysis of crosslinks of wt A3C or W74A in total cell lysate of transfected
cells. Cleared cell lysate was treated with 50mM NEM for 2hrs at room temperature and separated
on 4-12% SDS-PAGE. A3C wt or mutant were detected by an anti(α)-HA antibody.
Danksagung
DANKE… …Carsten für Deine großartige Unterstützung in dem Projekt, die Antworten auf (fast) alle Fragen,
Dein unendliches Wissen über Viren, das hervorragende Arbeitsklima, Deine IMMER offene Tür,
die Möglichkeit an so vielen Meetings teilzunehmen und, na klar, für 3C!
…Prof. Dr. Klaus Cichutek und Prof. Dr. Dieter Häussinger für die Möglichkeit in Ihren
Laborräumen und mit Ihren Mitteln forschen zu dürfen.
…Gerd und Arnulf dafür, dass ihr während der 4 Jahre immer Zeit gefunden habt, meine Daten
zu besprechen und kritisch zu beäugen, sowie für die unkomplizierte Übernahme der
Begutachtung dieser Arbeit.
…an die Lab-Münk Stamm-„11“:
…JZ für 4 Jahre als „Bench-Nachbar des Vertrauens“, Ironie, tolle Touri-Tage in New York, wilde
politische Theorien und Deine Hilfsbereitschaft. Und wenn’s ma wieder länger dauert – so en „Z-P“
hat uns nie geschadet!
…Mario für all Deine Hilfe und Ideen im Labor und Deine Gelassenheit. Hätte gerne öfter an
deinem „Kopfkino“ teilgehabt!
…Dani deine Herzlichkeit, Deinen Perfektionismus und das großartige Korrekturlesen! Ach ja, ich
werde Dein Tiramisu und Deine Kuchen vermissen!
…Gisbert und Benny für die sensationelle Zusammenarbeit und Euer sexy Modell!
…an alle „PEI-ler“: Andre, Andreas, Björn, Bodo, Egbert, Elea, Ferdi, Heide, Matthias, Julia,
Marion, Ralf, Stani, Sylvia und, last but not least, Frau Schmidt und Frau Varga.
…an alle „Labor-Düsseldorfer“: Anand, Andreas K, Andreas Pf, Benjamin, Björn, Boris, Carina,
Christian, Heiner, Katerina, Mariana, Melli, Saskia, Soraya, Ute und besonders der Wio.
…an die ODW - Giessen - Darmstadt - Düsseldorf - Buben und Mädels, die ihr mir immer
wieder zeigt, dass es mehr als Arbeit gibt! Danke für all die schönen und schlechten Stunden, die
Ihr mit mir geteilt habt. Es ist schön zu wissen, dass es Euch gibt, egal wo auf der Welt ihr euch
auch grad rumtreibt! Danke!
…Mama, Papa, Kerstin. Für alles!
Lebenslauf
Lebenslauf
Persönliche Daten Name: Henning Hofmann
geboren: 03. Februar 1981
in: Erbach/Odenwald
Schulausbildung 1986 – 1990 Grundschule, Bad König/Zell
1990 – 2000 Gymnasium Michelstadt
06/2000 Abitur
Universitätsausbildung 10/2000 – 09/2002 Grundstudium der Biologie an der Justus-Liebig Universität in
Giessen 09/2002 Vordiplom Biologie 10/2002 – 06/2006 Hauptstudium der Biologie an der Technischen Universität
Darmstadt 04/2005 – 12/2005 Anfertigung der Diplomarbeit unter Anleitung von Prof. Dr. H. Ulrich
Göringer am Institut für Mikrobiologie und Genetik der Technischen Universität Darmstadt
Titel: „TbRGG1 – ein oligo(U) und oligo(G) bindendes Protein aus
Trypanosoma brucei – Echtzeitstudien durch Oberflächen-resonanz“
06/2006 Diplom Biologie 08/2006 – 10/2008 Beginn der Anfertigung der Dissertation unter Anleitung von Dr.
Carsten Münk in der Abteilung Medizinische Biotechnologie des Paul-Ehrlich-Instituts in Langen
10/2008 – 08/2010 Weiterführung der begonnen Dissertation unter Anleitung von Prof.
Dr. Carsten Münk in der Klinik für Gastroenterologie, Hepatologie und Infektiologie an der Heinrich-Heine-Universität in Düsseldorf (Umzug der Arbeitsgruppe)
Titel: „The human cytidine deaminase APOBEC3C restricts retroviruses independent of editing - a biochemical and structural analysis”
Recommended